Latest YouTube Video

Saturday, November 26, 2016

Researchers Show How to Steal Tesla Car by Hacking into Owner's Smartphone

New technology is always a little scary, so are Smart Cars. From GPS system and satellite radio to wireless locks, steering, brakes, and accelerator, today vehicles are more connected to networks than ever, and so they are more hackable than ever. It's not new for security researchers to hack connected cars. Previously they had demonstrated how to hijack a car remotely, and how to disable


from The Hacker News http://ift.tt/2grcduh
via IFTTT

Beware! Malicious JPG Images on Facebook Messenger Spreading Locky Ransomware

If you receive an image file sent by someone, even your friend, on your Facebook Messenger, LinkedIn or any other social media platform, just DO NOT CLICK ON IT. Even JPG image file could eventually infect your computer with the infamous Locky Ransomware. Earlier this week, we reported a new attack campaign that used Facebook Messenger to spread Locky Ransomware via .SVG image files,


from The Hacker News http://ift.tt/2fOK4xj
via IFTTT

Apollo 17 VIP Site Anaglyph


Get out your red/blue glasses and check out this stereo scene from Taurus-Littrow valley on the Moon! The color anaglyph features a detailed 3D view of Apollo 17's Lunar Rover in the foreground -- behind it lies the Lunar Module and distant lunar hills. Because the world was going to be able to watch the Lunar Module's ascent stage liftoff via the rover's TV camera, this parking place was also known as the VIP Site. In December of 1972, Apollo 17 astronauts Eugene Cernan and Harrison Schmitt spent about 75 hours on the Moon, while colleague Ronald Evans orbited overhead. The crew returned with 110 kilograms of rock and soil samples, more than from any of the other lunar landing sites. Cernan and Schmitt are still the last to walk (or drive) on the Moon. via NASA http://ift.tt/2fyv60M

Friday, November 25, 2016

I have a new follower on Twitter


UserIQ
UserIQ is the first and only Customer Growth Platform™ that empowers SaaS companies to foster growth beyond the funnel.
Atlanta, GA
http://t.co/NsNSrVrIdo
Following: 854 - Followers: 1505

November 25, 2016 at 04:07PM via Twitter http://twitter.com/UserIQ

Ravens: LB Elvis Dumervil (foot), who has not played since Oct. 9, questionable for Sunday's game against the Bengals (ESPN)

from ESPN http://ift.tt/17lH5T2
via IFTTT

Ravens: CB Jimmy Smith (back) doubtful for Sunday's game against the Bengals (ESPN)

from ESPN http://ift.tt/17lH5T2
via IFTTT

pendo.io (@pendoio) liked one of your Tweets!

@mistermcguire: [FD] ISPconfig p6 - UI Exception & XSS Vulnerability   pendo.io liked your Tweet. View   Patrick McGuire @mistermcguire =   [FD] ISPconfig v3.0.5.4 p6 - UI Exception & XSS Vulnerability ift.tt/2bA4F9P   Settings | Help | Opt-out | Download app Twitter, Inc. 1355 Market Street, Suite 900 San Francisco, CA 94103

Source: Gmail -> IFTTT-> Blogger

I have a new follower on Twitter


Billy Soden
Always a teacher, and more so a learner. Dreamer, musician lifelong learner, and hope dealer. The most amazing people in the world call me Dad.
USA
https://t.co/rRQ63DbMFt
Following: 13196 - Followers: 12077

November 25, 2016 at 10:12AM via Twitter http://twitter.com/sodenbsoden

Live chat text strikethrough, no notification (anonymous user)

Live chat text strikethrough, no notification (anonymous user) #5048. Open. brylie opened this Issue 18 minutes ago · 0 comments ...

from Google Alert - anonymous http://ift.tt/2gnAvVW
via IFTTT

I have a new follower on Twitter


TTI Wireless
THE Leader for wired and wireless networks, data security and wireless communications.
Sayreville, NJ
http://t.co/wWHOdLnN0o
Following: 3832 - Followers: 4381

November 25, 2016 at 07:57AM via Twitter http://twitter.com/ttiwireless

[FD] UCanCode multiple vulnerabilities

http://ift.tt/2gn8uO7

Source: Gmail -> IFTTT-> Blogger

[FD] NEW VMSA-2016-0021 VMware product updates address partial information disclosure vulnerability

?

Source: Gmail -> IFTTT-> Blogger

[FD] NEW VMSA-2016-0022 VMware product updates address information disclosure vulnerabilities



Source: Gmail -> IFTTT-> Blogger

[FD] [SYSS-2016-107] EASY HOME Alarmanlagen-Set - Cryptographic Issues (CWE-310)

-----BEGIN PGP SIGNED MESSAGE-

Source: Gmail -> IFTTT-> Blogger

[FD] [SYSS-2016-106] EASY HOME Alarmanlagen-Set - Missing Protection against Replay Attacks

-----BEGIN PGP SIGNED MESSAGE-

Source: Gmail -> IFTTT-> Blogger

[FD] [SYSS-2016-072] Olypmia Protect 9061 - Missing Protection against Replay Attacks

-----BEGIN PGP SIGNED MESSAGE-

Source: Gmail -> IFTTT-> Blogger

[FD] [SYSS-2016-071] Blaupunkt Smart GSM Alarm SA 2500 Kit - Missing Protection against Replay Attacks

-----BEGIN PGP SIGNED MESSAGE-

Source: Gmail -> IFTTT-> Blogger

[FD] [SYSS-2016-066] Multi Kon Trade M2B GSM Wireless Alarm System - Missing Protection against Replay Attacks

-----BEGIN PGP SIGNED MESSAGE-

Source: Gmail -> IFTTT-> Blogger

[FD] [SYSS-2016-064] Multi Kon Trade M2B GSM Wireless Alarm System - Improper Restriction of Excessive Authentication Attempts (CWE-307)

-----BEGIN PGP SIGNED MESSAGE-

Source: Gmail -> IFTTT-> Blogger

[FD] Red Hat JBoss EAP deserialization of untrusted data

[FD] Faraday v2.2: Collaborative Penetration Test and Vulnerability Management Platform

Faraday is the Integrated Multiuser Risk Environment you were looking for! It maps and leverages all the knowledge you generate in real time, letting you track and understand your audits. Our dashboard for CISOs and managers uncovers the impact and risk being assessed by the audit in real-time without the need for a single email. Developed with a specialized set of functionalities that help users improve their own work, the main purpose is to re-use the available tools in the community taking advantage of them in a collaborative way! Check out the Faraday project in Github. http://ift.tt/1D8gKXz This release features a brand new library to connect with Faraday Server! Managing vulnerabilities is now easier in Faraday! Changes: - Status and creator fields: A simple change can go a long way - we added two new ways of classifying issues stored in Faraday. With the new update it is now possible to check the status of an issue - this could be opened, closed, re-opened or the risk is accepted. If you set a vulnerability status as closed and later on when you re-scan the target the same issue is found again, the status will automatically change into re-opened allowing you to have a more granular view of the results of your scans. This is perfect for doing remediation retests, helping you to quickly understand what is still vulnerable. Also, issues created by a specific tool, can now be filtered and sorted out. A great way to see where are the sources of information used during an engagement. For example, as we can see in the following screenshots, we have three different issues that are closed [1]. After we import a Nessus scan the issues are marked as re-opened [2], indicating that the vulnerability is still present in the last scan. - Added a message to configure the Webshell - added a descriptive message for users who don't have the Webshell properly configured - Fixed typo in Executive Report modal - New library to connect with Faraday Server - Fixed Fplugin, now it uses the new library to communicate with the Server - Refactor in Faraday Core and GTK Client - Bug fixing in Faraday Client and Server - News boxes example in the WEB UI - New plugins: Dirb, Netdiscover, FruityWifi, Sentinel - Improvements on the WPscan plugin - Fixed Licenses search - there was a bug that disabled the option to search for licenses, now it is fixed and full-text search is enabled in the Licenses component - Refactor Licenses module to be compatible with JS Strict Mode - in our efforts to improve our existing codebase for the WEB UI we refactored this component in order to make it run using Strict Mode in JavaScript We hope you enjoy it, and let us know if you have any questions or comments. http://ift.tt/1D4inIk http://ift.tt/1D8gKXz https://twitter.com/faradaysec http://ift.tt/2cWL046

Source: Gmail -> IFTTT-> Blogger

[FD] [CVE-2016-7098] GNU Wget < 1.18 Access List Bypass / Race Condition

Vulnerability: GNU Wget < 1.18 Access List Bypass / Race Condition CVE-2016-7098 Discovered by: Dawid Golunski (@dawid_golunski) http://ift.tt/2fcYckq Severity: Medium GNU wget in version 1.17 and earlier, when used in mirroring/recursive mode, is affected by a Race Condition vulnerability that might allow remote attackers to bypass intended wget access list restrictions specified with -A parameter. This might allow attackers to place malicious/restricted files onto the system. Depending on the application / download directory, this could potentially lead to other vulnerabilities such as code execution etc. The full advisory and a PoC exploit can be found at: http://ift.tt/2fHADQn For updates, follow: https://twitter.com/dawid_golunski

Source: Gmail -> IFTTT-> Blogger

[FD] The HS-110 Smart Plug aka Projekt Kasa

Content Table 1. Introduction 2. The Firmware 3. The Android Application 4. The Problems 5. Conclusion 6. Appendix 6.1. Excursion Dalvik 6.2 Control script 1. Introduction The HS-110 is a Smart Plug meaning it is capable of being controlled with commands via a network. TP-Link released a mobile application called "Kasa for Mobile" for Android and iOS devices to control the Smart Plug. The possibilities range from simple tasks like turning the Plug on and off to advanced options like planing schedules and timers. The HS-110 additionally has the possibility to measure and store data regarding power consumption. These are screenshots of the app home screen, the main control and the settings for a plug: app control screen plug control screen plug settings The device itself is pretty straightforward with only two buttons. The one at the top is the reset button and the other one in the front is the power button and status led: plug from the front plug from the top plug from the back To open it we remove the hidden screw under the information sheet and then break it open using a little bit of force: [open1] [open2] Now we remove the top part of the board and the two screws on the second part to get rid of the plastic hull: [open3] [open4] [open5] We can now see the Atheros AR9331 (Hornet) on the right board in the middle picture above. It is a System-on-a-Chip (SOC) which has a MIPS 24K processor and is a full featured IEEE 802.11n 1x1 AP/Router. It also has a 32 MiB RAM (Zentel A3S56D40GTP-50l) on the opposite side of the same board. The other board hosts the electronics for the actual plug. But the interesting question is: What this SOC is actually running so let's move on to the next section. 2. The Firmware The Smart Plug runs on a 64-bit Linux (2.6.31). The Firmware is available at the Website of TP-Link. Our version is 1.0.7. There is also an unofficial unstable API on GitHub. For a first analysis of the Firmware we used binwalk . It is important to also install sasquatch for this since unsquashfs appears to have issues with TP-Link firmware. You can just install the necessary tools for the installation of sasquatch via apt sudo apt-get install build-essential liblzma-dev liblzo2-dev zlib1g-dev or the corresponding packages if you don't use apt. After that just clone the sasquatch git repository and run the build script. At the end we have to install binwalk by cloning it's git repository and running the setup.py script via sudo python setup.py install or sudo python3 setup.py install if you are using python3.x. For the dependencies we can run deps.sh, at least when we are using apt. Otherwise you have to install them by yourself. A list is available at github . Now we are ready to run binwalk at the firmware with following command: root@kali:~/Desktop/test# binwalk hs110v1_us_1.0.7_Build_151016_Rel.24186.bin DECIMAL HEXADECIMAL DESCRIPTION

Source: Gmail -> IFTTT-> Blogger

[FD] CVE-2013-3120 MSIE 10 MSHTML CEditAdorner::Detach use-after-free details

Throughout November, I plan to release details on vulnerabilities I found in web-browsers which I've not released before. This is the nineteenth entry in that series. Unfortunately I won't be able to publish everything within one month at the current rate, so I may continue to publish these through December and January. The below information is available in more detail on my blog at http://ift.tt/2fYVIIU. There you can find a repro that triggered this issue in addition to the information below. Today's release is a bit skimpy as it was found over 3 years ago, back when I did not have the tool-set to analyze issues that I have today. Follow me on http://twitter.com/berendjanwever for daily browser bugs. MSIE 10 MSHTML CEditAdorner::Detach use-after-free ================================================== (MS13-047, CVE-2013-3120) Synopsis

Source: Gmail -> IFTTT-> Blogger

[FD] Microsoft Internet Explorer 11 MSHTML CGeneratedContent::HasGeneratedSVGMarker type confusion

Throughout November, I plan to release details on vulnerabilities I found in web-browsers which I've not released before. This is the eighteenth entry in that series. Unfortunately I won't be able to publish everything within one month at the current rate, so I may continue to publish these through December and January. The below information is available in more detail on my blog at http://ift.tt/2fLRgdr. There you can find a repro that triggered this issue, reversed snippets from the vulnerable code and more details in addition to the information below. Follow me on http://twitter.com/berendjanwever for daily browser bugs. MSIE 11 MSHTML CGeneratedContent::HasGeneratedSVGMarker type confusion ======================================================= (The fix and CVE number for this issue are unknown) Synopsis

Source: Gmail -> IFTTT-> Blogger

[FD] CVE-2015-1251: Chrome blink Speech­Recognition­Controller use-after-free details

Throughout November, I plan to release details on vulnerabilities I found in web-browsers which I've not released before. This is the seventeenth entry in that series. Unfortunately I won't be able to publish everything within one month at the current rate, so I may continue to publish these through December and January. The below information is available in more detail on my blog at http://ift.tt/2gAt9Tk. There you can find a repro that triggered this issue, snippets from the vulnerable source and more details in addition to the information below. Follow me on http://twitter.com/berendjanwever for daily browser bugs. Chrome blink SpeechRecognitionController use-after-free ======================================================= (Chromium 455857, CVE-2015-1251) Synopsis

Source: Gmail -> IFTTT-> Blogger

[FD] CVE-2015-0050: Microsoft Internet Explorer 8 MSHTML SRunPointer::SpanQualifier/RunType OOB read details

Throughout November, I plan to release details on vulnerabilities I found in web-browsers which I've not released before. This is the sixteenth entry in that series. Unfortunately I won't be able to publish everything within one month at the current rate, so I may continue to publish these through December and January. The below information is available in more detail on my blog at http://ift.tt/2gg0Mal. There you can find a repro that triggered this issue in addition to the information below. Follow me on http://twitter.com/berendjanwever for daily browser bugs. MSIE 8 MSHTML SRunPointer::SpanQualifier/RunType OOB read ========================================================= (MS15-009, CVE-2015-0050) Synopsis

Source: Gmail -> IFTTT-> Blogger

[FD] MobSF v0.9.3 is Released: Now supports Windows APPX Static Analysis

Hello Folks, MobSF v0.9.3 is released. About MobSF Mobile Security Framework (MobSF) is an intelligent, all-in-one open source mobile application (Android/iOS/Windows) automated pen-testing framework capable of performing static and dynamic analysis. It can be used for effective and fast security analysis of Android, iOS and Windows mobile Applications and supports both binaries (APK, IPA & APPX ) and zipped source code. MobSF can also perform Web API Security testing with it's API Fuzzer that can do Information Gathering, analyze Security Headers, identify Mobile API specific vulnerabilities like XXE, SSRF, Path Traversal, IDOR, and other logical issues related to Session and API Rate Limiting. What's new in v0.9.3 * Clipboard Monitor for Android Dynamic Analysis * Windows APPX Static Analysis Support * Added Docker File * Added Support for Kali Linux * Code Quality and Lintering * Partial PEP8 Formating, Code Refactoring and Restructuring * Imporved Static Analyzer Regex * Disabling Syntax Highlighter Edit mode * More MIME Type additions * Update File Upload Size to 100 MB * MobSFfy script to support commandline args * New strings.py tool for string extraction in iOS Apps. * Updated iOS Static Analysis ruleset. * Django Upgrade to 1.10 * MobSF VM 0.3 Released Download v0.9.3: http://ift.tt/2fZ71ys GitHub: http://ift.tt/1Tj7oyN Documentation: http://ift.tt/1Mhye89 Regards, Ajin Abraham, Security Enthusiast, Bangalore, India +91-9633325997 Linkedin: http://ift.tt/1P92eTn Twitter: https://twitter.com/ajinabraham Web: http://opensecurity.in | https://opsecx.com

Source: Gmail -> IFTTT-> Blogger

Thursday, November 24, 2016

I have a new follower on Twitter


hulu
Bond is now on Hulu. Oh, fine. Bond... James Bond is now on Hulu. Need help? Tweet @hulu_support
Los Angeles, CA, USA
https://t.co/vTv6CeaZbx
Following: 31498 - Followers: 262432

November 24, 2016 at 09:42PM via Twitter http://twitter.com/hulu

I have a new follower on Twitter


MBOSSTRADE
MARTINGALE BINARY OPTIONS SIGNALS SERVICE 📱 Real-time Signals via Whatsapp 💹 Register with our broker and enjoy 🆓 membership
England, United Kingdom
https://t.co/TTx1SQg4si
Following: 1252 - Followers: 615

November 24, 2016 at 05:30PM via Twitter http://twitter.com/mbosstrade1

Microsoft Shares Telemetry Data Collected from Windows 10 Users with 3rd-Party

Cyber security is a major challenge in today's world, as cyber attacks have become more automated and difficult to detect, where traditional cyber security practices and systems are no longer sufficient to protect businesses, governments, and other organizations. In past few years, Artificial Intelligence and Machine Learning had made a name for itself in the field of cyber security, helping


from The Hacker News http://ift.tt/2g8JhYN
via IFTTT

I have a new follower on Twitter


pendo.io
Pendo delivers product success. Understand and guide the complete user journey through your application with analytics, user feedback, & contextual help.
United States
http://t.co/fZUIdIoxzd
Following: 16400 - Followers: 17466

November 24, 2016 at 12:09PM via Twitter http://twitter.com/pendoio

[FD] [RT-SA-2016-003] Less.js: Compilation of Untrusted LESS Files May Lead to Code Execution through the JavaScript Less Compiler

Advisory: Less.js: Compilation of Untrusted LESS Files May Lead to Code Execution through the JavaScript Less Compiler RedTeam Pentesting discovered behaviour in the Less.js compiler, which allows execution of arbitrary code if an untrusted LESS file is compiled. Details ======= Product: Less Compiler Affected Versions: probably all versions Fixed Versions: none Vulnerability Type: Code Execution Security Risk: low Vendor URL: http://lesscss.org/ Vendor Status: decided not to fix Advisory URL: http://ift.tt/2gk77jd Advisory Status: published Introduction ============ "Less is a CSS pre-processor, meaning that it extends the CSS language, adding features that allow variables, mixins, functions and many other techniques that allow you to make CSS that is more maintainable, themable and extendable. Less runs inside Node, in the browser and inside Rhino. There are also many 3rd party tools that allow you to compile your files and watch for changes." (from the project's homepage) More Details ============ The Less project provides a compiler [0] to transform LESS code into CSS. Among other features, it supports embedded inline JavaScript code in LESS files. To our knowledge, this feature is currently not mentioned in the official documentation provided by the Less project. However, while researching the history of the Less website it was discovered that this feature was indeed documented in the past [1]. Third parties also document this feature [2]. The following example shows how this feature can be used. JavaScript code can be embedded in LESS by enclosing it in backticks. In the following, the result of the expression '1+1' is assigned to the variable 'test':

Source: Gmail -> IFTTT-> Blogger

THN Deal — Learn Wi-Fi Hacking & Penetration Testing [Online Course: 83% OFF]

Hacking Wi-Fi is not a trivial process, but it does not take too long to learn. If you want to learn WiFi Hacking and Penetration testing, you are at right place. Don't associate hacking as a negative, as you can learn some hacking skills yourself to secure your networks and devices. WiFi hacking is an all time hot topic among hackers as well as penetration testers. This week's featured deal


from The Hacker News http://ift.tt/2gk0CwO
via IFTTT

Antivirus Firm Kaspersky launches Its Own Secure Operating System

The popular cyber security and antivirus company Kaspersky has unveiled its new hack-proof operating system: Kaspersky OS. The new operating system has been in development for last 14 years and has chosen to design from scratch rather than relying on Linux. Kaspersky OS makes its debut on a Kraftway Layer 3 Switch, CEO Eugene Kaspersky says in his blog post, without revealing many details


from The Hacker News http://ift.tt/2gjxs14
via IFTTT

Binary options anonymous

Long-playing next Boniface dissolvings anonymous covenanter binary options anonymous guillotining truck offhanded? Spondylitic Piggy adjuring ...

from Google Alert - anonymous http://ift.tt/2fVcXLj
via IFTTT

FBI Hacked into 8,000 Computers in 120 Countries Using A Single Warrant

The FBI hacked into more than 8,000 computers in 120 different countries with just a single warrant during an investigation into a dark web child pornography website, according to a newly published court filings. This FBI's mass hacking campaign is related to the high-profile child pornography Playpen case and represents the largest law enforcement hacking campaign known to date. The warrant


from The Hacker News http://ift.tt/2fUJpgH
via IFTTT

NGC 7635: Bubble in a Cosmic Sea


Do you see the bubble in the center? Seemingly adrift in a cosmic sea of stars and glowing gas, the delicate, floating apparition in this widefield view is cataloged as NGC 7635 - The Bubble Nebula. A mere 10 light-years wide, the tiny Bubble Nebula and the larger complex of interstellar gas and dust clouds are found about 11,000 light-years distant, straddling the boundary between the parental constellations Cepheus and Cassiopeia. Also included in the breathtaking vista is open star cluster M52 (upper left), some 5,000 light-years away. The featured image spans about two degrees on the sky corresponding to a width of about 375 light-years at the estimated distance of the Bubble Nebula. via NASA http://ift.tt/2gkaqJc

Wednesday, November 23, 2016

Feature Importance Measure for Non-linear Learning Algorithms. (arXiv:1611.07567v1 [cs.AI])

Complex problems may require sophisticated, non-linear learning methods such as kernel machines or deep neural networks to achieve state of the art prediction accuracies. However, high prediction accuracies are not the only objective to consider when solving problems using machine learning. Instead, particular scientific applications require some explanation of the learned prediction function. Unfortunately, most methods do not come with out of the box straight forward interpretation. Even linear prediction functions are not straight forward to explain if features exhibit complex correlation structure.

In this paper, we propose the Measure of Feature Importance (MFI). MFI is general and can be applied to any arbitrary learning machine (including kernel machines and deep learning). MFI is intrinsically non-linear and can detect features that by itself are inconspicuous and only impact the prediction function through their interaction with other features. Lastly, MFI can be used for both --- model-based feature importance and instance-based feature importance (i.e, measuring the importance of a feature for a particular data point).



from cs.AI updates on arXiv.org http://ift.tt/2g5J9cw
via IFTTT

Programs as Black-Box Explanations. (arXiv:1611.07579v1 [stat.ML])

Recent work in model-agnostic explanations of black-box machine learning has demonstrated that interpretability of complex models does not have to come at the cost of accuracy or model flexibility. However, it is not clear what kind of explanations, such as linear models, decision trees, and rule lists, are the appropriate family to consider, and different tasks and models may benefit from different kinds of explanations. Instead of picking a single family of representations, in this work we propose to use "programs" as model-agnostic explanations. We show that small programs can be expressive yet intuitive as explanations, and generalize over a number of existing interpretable families. We propose a prototype program induction method based on simulated annealing that approximates the local behavior of black-box classifiers around a specific prediction using random perturbations. Finally, we present preliminary application on small datasets and show that the generated explanations are intuitive and accurate for a number of classifiers.



from cs.AI updates on arXiv.org http://ift.tt/2g5HnYF
via IFTTT

Efficient Delivery Policy to Minimize User Traffic Consumption in Guaranteed Advertising. (arXiv:1611.07599v1 [cs.DS])

In this work, we study the guaranteed delivery model which is widely used in online display advertising. In the guaranteed delivery scenario, ad exposures (which are also called impressions in some works) to users are guaranteed by contracts signed in advance between advertisers and publishers. A crucial problem for the advertising platform is how to fully utilize the valuable user traffic to generate as much as possible revenue.

Different from previous works which usually minimize the penalty of unsatisfied contracts and some other cost (e.g. representativeness), we propose the novel consumption minimization model, in which the primary objective is to minimize the user traffic consumed to satisfy all contracts. Under this model, we develop a near optimal method to deliver ads for users. The main advantage of our method lies in that it consumes nearly as least as possible user traffic to satisfy all contracts, therefore more contracts can be accepted to produce more revenue. It also enables the publishers to estimate how much user traffic is redundant or short so that they can sell or buy this part of traffic in bulk in the exchange market. Furthermore, it is robust with regard to priori knowledge of user type distribution. Finally, the simulation shows that our method outperforms the traditional state-of-the-art methods.



from cs.AI updates on arXiv.org http://ift.tt/2gCwkK7
via IFTTT

Multi-Modal Mean-Fields via Cardinality-Based Clamping. (arXiv:1611.07941v1 [cs.CV])

Mean Field inference is central to statistical physics. It has attracted much interest in the Computer Vision community to efficiently solve problems expressible in terms of large Conditional Random Fields. However, since it models the posterior probability distribution as a product of marginal probabilities, it may fail to properly account for important dependencies between variables. We therefore replace the fully factorized distribution of Mean Field by a weighted mixture of such distributions, that similarly minimizes the KL-Divergence to the true posterior. By introducing two new ideas, namely, conditioning on groups of variables instead of single ones and using a parameter of the conditional random field potentials, that we identify to the temperature in the sense of statistical physics to select such groups, we can perform this minimization efficiently. Our extension of the clamping method proposed in previous works allows us to both produce a more descriptive approximation of the true posterior and, inspired by the diverse MAP paradigms, fit a mixture of Mean Field approximations. We demonstrate that this positively impacts real-world algorithms that initially relied on mean fields.



from cs.AI updates on arXiv.org http://ift.tt/2gCwp0i
via IFTTT

On Design Mining: Coevolution and Surrogate Models. (arXiv:1506.08781v6 [cs.NE] UPDATED)

Design mining is the use of computational intelligence techniques to iteratively search and model the attribute space of physical objects evaluated directly through rapid prototyping to meet given objectives. It enables the exploitation of novel materials and processes without formal models or complex simulation. In this paper, we focus upon the coevolutionary nature of the design process when it is decomposed into concurrent sub-design threads due to the overall complexity of the task. Using an abstract, tuneable model of coevolution we consider strategies to sample sub-thread designs for whole system testing and how best to construct and use surrogate models within the coevolutionary scenario. Drawing on our findings, the paper then describes the effective design of an array of six heterogeneous vertical-axis wind turbines.



from cs.AI updates on arXiv.org http://ift.tt/1Nx7TnZ
via IFTTT

A Brain-like Cognitive Process with Shared Methods. (arXiv:1507.04928v4 [cs.AI] UPDATED)

This paper describes a detailed cognitive structure with related processes. The system is modelled on the human brain, where pattern creation and activation would be automatic and distributed. Methods for carrying out the processes are also suggested. The main purpose of this paper is to reaffirm earlier research on different knowledge-based and experience-based clustering techniques. The overall architecture has stayed essentially the same and so it is the localised processes or smaller details that have been updated. For example, a counting mechanism is used slightly differently, to measure a level of 'cohesion' instead of a 'correct' classification, over pattern instances. The introduction of features has further enhanced the architecture and a new entropy-style of equation is proposed. While an earlier paper defined three levels of functional requirement, this paper re-defines the levels in a more human vernacular, with higher-level goals described in terms of action-result pairs.



from cs.AI updates on arXiv.org http://ift.tt/1LlCiaY
via IFTTT

Beyond knowing that: a new generation of epistemic logics. (arXiv:1605.01995v3 [cs.AI] UPDATED)

Epistemic logic has become a major field of philosophical logic ever since the groundbreaking work by Hintikka (1962). Despite its various successful applications in theoretical computer science, AI, and game theory, the technical development of the field has been mainly focusing on the propositional part, i.e., the propositional modal logics of "knowing that". However, knowledge is expressed in everyday life by using various other locutions such as "knowing whether", "knowing what", "knowing how" and so on (knowing-wh hereafter). Such knowledge expressions are better captured in quantified epistemic logic, as was already discussed by Hintikka (1962) and his sequel works at length. This paper aims to draw the attention back again to such a fascinating but largely neglected topic. We first survey what Hintikka and others did in the literature of quantified epistemic logic, and then advocate a new quantifier-free approach to study the epistemic logics of knowing-wh, which we believe can balance expressivity and complexity, and capture the essential reasoning patterns about knowing-wh. We survey our recent line of work on the epistemic logics of "knowing whether", "knowing what" and "knowing how" to demonstrate the use of this new approach.



from cs.AI updates on arXiv.org http://ift.tt/24EIYeE
via IFTTT

Synthesizing the preferred inputs for neurons in neural networks via deep generator networks. (arXiv:1605.09304v5 [cs.NE] UPDATED)

Deep neural networks (DNNs) have demonstrated state-of-the-art results on many pattern recognition tasks, especially vision classification problems. Understanding the inner workings of such computational brains is both fascinating basic science that is interesting in its own right - similar to why we study the human brain - and will enable researchers to further improve DNNs. One path to understanding how a neural network functions internally is to study what each of its neurons has learned to detect. One such method is called activation maximization (AM), which synthesizes an input (e.g. an image) that highly activates a neuron. Here we dramatically improve the qualitative state of the art of activation maximization by harnessing a powerful, learned prior: a deep generator network (DGN). The algorithm (1) generates qualitatively state-of-the-art synthetic images that look almost real, (2) reveals the features learned by each neuron in an interpretable way, (3) generalizes well to new datasets and somewhat well to different network architectures without requiring the prior to be relearned, and (4) can be considered as a high-quality generative method (in this case, by generating novel, creative, interesting, recognizable images).



from cs.AI updates on arXiv.org http://ift.tt/1U8Wg8k
via IFTTT

VIME: Variational Information Maximizing Exploration. (arXiv:1605.09674v3 [cs.LG] UPDATED)

Scalable and effective exploration remains a key challenge in reinforcement learning (RL). While there are methods with optimality guarantees in the setting of discrete state and action spaces, these methods cannot be applied in high-dimensional deep RL scenarios. As such, most contemporary RL relies on simple heuristics such as epsilon-greedy exploration or adding Gaussian noise to the controls. This paper introduces Variational Information Maximizing Exploration (VIME), an exploration strategy based on maximization of information gain about the agent's belief of environment dynamics. We propose a practical implementation, using variational inference in Bayesian neural networks which efficiently handles continuous state and action spaces. VIME modifies the MDP reward function, and can be applied with several different underlying RL algorithms. We demonstrate that VIME achieves significantly better performance compared to heuristic exploration methods across a variety of continuous control tasks and algorithms, including tasks with very sparse rewards.



from cs.AI updates on arXiv.org http://ift.tt/1RKQdoC
via IFTTT

Deep Reinforcement Learning for Robotic Manipulation with Asynchronous Off-Policy Updates. (arXiv:1610.00633v2 [cs.RO] UPDATED)

Reinforcement learning holds the promise of enabling autonomous robots to learn large repertoires of behavioral skills with minimal human intervention. However, robotic applications of reinforcement learning often compromise the autonomy of the learning process in favor of achieving training times that are practical for real physical systems. This typically involves introducing hand-engineered policy representations and human-supplied demonstrations. Deep reinforcement learning alleviates this limitation by training general-purpose neural network policies, but applications of direct deep reinforcement learning algorithms have so far been restricted to simulated settings and relatively simple tasks, due to their apparent high sample complexity. In this paper, we demonstrate that a recent deep reinforcement learning algorithm based on off-policy training of deep Q-functions can scale to complex 3D manipulation tasks and can learn deep neural network policies efficiently enough to train on real physical robots. We demonstrate that the training times can be further reduced by parallelizing the algorithm across multiple robots which pool their policy updates asynchronously. Our experimental evaluation shows that our method can learn a variety of 3D manipulation skills in simulation and a complex door opening skill on real robots without any prior demonstrations or manually designed representations.



from cs.AI updates on arXiv.org http://ift.tt/2dDMPQm
via IFTTT

Anonymous read access to a project

My proposal is to add a new possibility for the "invite by email" feature, which would give the user a read-only anonymous access. Typically, they ...

from Google Alert - anonymous http://ift.tt/2gClTpU
via IFTTT

anonymous users seeing empty results page, with no URL parameter passed after using /search/site

As an anonymous user, if you go to http://ift.tt/2gilh8p and type in a valid word. You get an empty results page. With the normal helper ...

from Google Alert - anonymous http://ift.tt/2fGFykr
via IFTTT

Ravens: Steve Smith jokes about "losing sleep" over rookie CBs saying they don't respect him; "they don't really get it" (ESPN)

from ESPN http://ift.tt/17lH5T2
via IFTTT

[FD] Stored Cross-Site Scripting in Gallery - Image Gallery WordPress Plugin

--------------------------------------------------------------------

Source: Gmail -> IFTTT-> Blogger

Your Headphones Can Spy On You — Even If You Have Disabled Microphone

Have you considered the possibility that someone could be watching you through your webcam? Or Listening to all your conversations through your laptop’s microphone? Even a bit of thought about this probability could make you feel incredibly creepy. But most people think that they have a solution to these major issues i.e. simply covering your laptop’s webcam and microphone, just like


from The Hacker News http://ift.tt/2gggcd7
via IFTTT

ISS Daily Summary Report – 11/22/2016

Waste and Hygiene Compartment (WHC) Maintenance:  Ground specialists met to discuss the WHC leak that occurred yesterday.  Based on review of the imagery the crew took during the cleanup, the ground specialists determined the most likely source of the leak to be a hydroconnector on the Urine Valve Block, but determined that since all of the contaminated hardware would need to be replaced there was not a significant benefit to performing troubleshooting to isolate the root cause of the leak.  Today, the crew replaced the Pump Separator, Urine Valve Block, several hoses and a data cable.  The WHC was activated, checked out and is go for nominal operations. Fluid Shifts Before, During and After Prolonged Space Flight and Their Association with Intracranial Pressure and Visual Impairment (Fluid Shifts): Today, a 48S crewmember began the first part of the Flight Day 45 (FD-45) Fluid Shifts data collection.  Upon wakeup he collected baseline saliva, urine, and with another crewmember acting as operator collected blood samples that were inserted into Minus Eighty Degree Celsius Laboratory Freezer for ISS (MELFI) prior to ingesting a Sodium Bromide (NaBr) tracer.  Throughout the day, the crew will perform more urine, blood, and saliva collections, inserting those samples into MELFI as well.  Fluid Shifts investigates the causes for severe and lasting physical changes to astronaut’s eyes. Because the headward fluid shift is a hypothesized contributor to these changes, reversing this fluid shift with a lower body negative pressure device is investigated as a possible intervention. Results from this study may help to develop preventative measures against lasting changes in vision and eye damage. Water Monitoring Suite (WMS) and Aquapad Sampling: In coordination with regular Environmental Health System (EHS) water sampling, the crew is scheduled to perform the ESA sponsored Aquapad (Paper Analytical Device) technological demonstration and sampling using the Microbial Monitoring System (MMS) portion of the WMS experiment.  By using the same water collected for the EHS sample, ground teams can evaluate the capabilities of the new hardware.  For the MMS, the crew reviewed an On-Board Training (OBT) video then configured the hardware to sample a low and high DNA concentration sample using the Razor Polymerase Chain Reaction (PCR) Microbial Monitoring System.  The WMS is a set of hardware that monitors microbes, silica and organic material in the water supply on the ISS. The hardware ensures crew members can test and monitor the safety of their water supplies on future space missions, especially on long-duration missions to Mars, asteroids or other destinations where Earth-based testing would be difficult or impossible. Aquapad is a new approach developed by France’s CNES space agency: paper impregnated with powdered growth medium creates a 3D petri dish. When water is added, the microbes form colored spots revealing their locations. Using a tablet computer application the crew will photograph the dots to calculate precisely how many bacteria are present.  GeoFlow-2: The crew activated the Multi-purpose Computer and Communications (MPCC) laptop, Fluid Science Laboratory (FSL) rack, installed Anti Vibration Mount (AVM) brackets then released the Facility Core Element in preparation for free-floating microgravity science inside the FSL.  Geoflow-2 studies heat and fluid flow currents within the Earth’s mantle. Geoflow-2 aims to improve computational methods that scientists and engineers use to understand and predict the processes in the Earth’s mantle that lead to volcanic eruptions, plate tectonics and earthquakes. Spacecraft Fire Safety (Saffire)-II: Operations started onboard Cygnus about 2325 GMT yesterday and wrapped up about 0230 GMT when the post-test flow visualization completed. Preliminary telemetry indicated that data and images were recorded as expected for all 9 samples and flow visualization. The Saffire-II hardware performed nominally with no apparent issues.  The team is downlinking the sensor and image data and begin analysis of the dataset later today. Japanese Experiment Module (JEM) Stowage Wire Kit Installation:  The crew started the activities to install the JEM Stowage Wire Kit.  This Kit will make additional stowage available in the area under the JEM Airlock (JEMAL).  The crew completed the activity to gather the hardware, but the rest of the activity was deferred due to work on the WHC. Emergency Role/Responsibility Review:  The crew performed a review of the Emergency General Instructions for response to emergencies and roles of each crew member.  This review included a discussion of roles of Soyuz Commander, ISS Commander and each crewmember and strategies for working through the response to Emergencies.  Mobile Servicing System (MSS) Operations:  The Robotics Ground Controllers translated the Mobile Transporter (MT) from Worksite (WS) 6 to WS7.  Controllers will walk-off the Space Station Remote Manipulator System (SSRMS) from Mobile Base System (MBS) 3 to the MBS 1 grapple fixture.  They will then unstow Special Purpose Dexterous Manipulator (SPDM).  These activities are in preparation for Robotic External Leak Locator (RELL) operations planned to start Monday 28-November. Today’s Planned Activities All activities were completed unless otherwise noted. Fluid Shifts Saliva Baseline Collection Fluid Shifts Urine Collection Fluid Shifts MELFI Insertion Operations Fluid Shifts Blood Baseline Collection Fluid Shifts Galley Water Collection Tracer Ingestion Operations Fluid Shifts Refrigerated Centrifuge Configuration KORREKTSIYA. Logging Liquid and Food (Medicine) Intake Fluid Shifts USOS Operations Historical Documentation Photography Virus Definition File Update on Auxiliary Computer System (ВКС) Laptops FSL MPCC Laptop Hardware Relocation and Powering on Laptop In Flight Maintenance Waste and Hygiene Compartment Stowage Gather. Fluid Shifts MELFI Insertion БД-2 R&R Ops. Potable Water Dispenser (PWD) Sample Collect FSL AVM brackets installation & Facility Core Element release in preparation of scientific operations Fluid Shifts Urine Collection – Subject WHC Urine Hydraulic Components R&R Electrostatic Levitation Furnace(ELF) Argon gas QD connection Total Organic Carbon Analyzer (TOCA) Potable Water Dispenser (PWD) Sample Analysis Fluid Shifts MELFI Retrieve Insertion Operations БМП Ф1 Absorption Cartridge Regeneration (start) JEM Common Gas Support Equipment Gas Supply Start Handover of Increment 49/50 Crew JEM Stowage Wire Kit Installation Preparation Fluid Shifts Blood 3-Hour Collection Fluid Shifts Refrigerated Centrifuge Configuration CEVIS Handover Video Review ПТАБ-1М repositioning (from pos.А306 to pos.А308). Photography of de-installed ПТАБ-1М (А308) CEVIS Crew Handover Fluid Shifts MELFI Retrieve Insertion Operations […]

from ISS On-Orbit Status Report http://ift.tt/2g3lIR0
via IFTTT

How to solve this problem? Assertion in void __stdcall

Assertion in void __stdcall `anonymous-namespace'::mwJavaAbort(void) at b:\matlab\src\jmi\javainit.cpp line 1356: Fatal Java Exception. See Java ...

from Google Alert - anonymous http://ift.tt/2g2XExA
via IFTTT

NTP DoS Exploit Released — Update Your Servers to Patch 10 Flaws

A proof-of-concept (PoC) exploit for a critical vulnerability in the Network Time Protocol daemon (ntpd) has been publically released that could allow anyone to crash a server with just a single maliciously crafted packet. The vulnerability has been patched by the Network Time Foundation with the release of NTP 4.2.8p9, which includes a total of 40 security patches, bug fixes, and


from The Hacker News http://ift.tt/2gjsNOb
via IFTTT

Plutos Sputnik Planum


Is there an ocean below Sputnik Planum on Pluto? The unusually smooth 1000-km wide golden expanse, visible in the featured image from New Horizons, appears segmented into convection cells. But how was this region created? One hypothesis now holds the answer to be a great impact that stirred up an underground ocean of salt water roughly 100-kilometers thick. The featured image of Sputnik Planum, part of the larger heart-shaped Tombaugh Regio, was taken last July and shows true details in exaggerated colors. Although the robotic New Horizons spacecraft is off on a new adventure, continued computer-modeling of this surprising surface feature on Pluto is likely to lead to more refined speculations about what lies beneath. via NASA http://ift.tt/2ghUTco

Massive Lightning Storm Lights up Northern Alabama

The Geostationary Operational Environmental Satellites - R Series (GOES-R) is the next generation of geostationary Earth-observing systems. The GOES-R program is a collaborative development and acquisition effort between the National Oceanic and Atmospheric Administration (NOAA) and NASA. The advanced spacecraft and instrument technology employed by the GOES-R series will provide significant improvements in the detection and observations of environmental phenomena that directly affect public safety, protection of property, and our nation's economic health and prosperity. The first satellite in the GOES-R series is scheduled for launch in October 2016. For more information on the GOES-R program please visit: goes-r.gov

from NASA's Scientific Visualization Studio: Most Recent Items http://ift.tt/2f5fl2C
via IFTTT

Tuesday, November 22, 2016

Lulu Jieyao (Anonymous)

Misc. Comments. "Lulu Jieyao, printed on blueprint paper in Chinese and Manchu, is stored in the National Beijing Library. There is no author's name.

from Google Alert - anonymous http://ift.tt/2fpiR6K
via IFTTT

An Efficient Training Algorithm for Kernel Survival Support Vector Machines. (arXiv:1611.07054v1 [cs.LG])

Survival analysis is a fundamental tool in medical research to identify predictors of adverse events and develop systems for clinical decision support. In order to leverage large amounts of patient data, efficient optimisation routines are paramount. We propose an efficient training algorithm for the kernel survival support vector machine (SSVM). We directly optimise the primal objective function and employ truncated Newton optimisation and order statistic trees to significantly lower computational costs compared to previous training algorithms, which require $O(n^4)$ space and $O(p n^6)$ time for datasets with $n$ samples and $p$ features. Our results demonstrate that our proposed optimisation scheme allows analysing data of a much larger scale with no loss in prediction performance. Experiments on synthetic and 5 real-world datasets show that our technique outperforms existing kernel SSVM formulations if the amount of right censoring is high ($\geq85\%$), and performs comparably otherwise.



from cs.AI updates on arXiv.org http://ift.tt/2f4NPCg
via IFTTT

A Deep Learning Approach for Joint Video Frame and Reward Prediction in Atari Games. (arXiv:1611.07078v1 [cs.AI])

Reinforcement learning is concerned with learning to interact with environments that are initially unknown. State-of-the-art reinforcement learning approaches, such as DQN, are model-free and learn to act effectively across a wide range of environments such as Atari games, but require huge amounts of data. Model-based techniques are more data-efficient, but need to acquire explicit knowledge about the environment dynamics or the reward structure.

In this paper we take a step towards using model-based techniques in environments with high-dimensional visual state space when system dynamics and the reward structure are both unknown and need to be learned, by demonstrating that it is possible to learn both jointly. Empirical evaluation on five Atari games demonstrate accurate cumulative reward prediction of up to 200 frames. We consider these positive results as opening up important directions for model-based RL in complex, initially unknown environments.



from cs.AI updates on arXiv.org http://ift.tt/2fnecyJ
via IFTTT

Interpreting Finite Automata for Sequential Data. (arXiv:1611.07100v1 [stat.ML])

Automaton models are often seen as interpretable models. Interpretability itself is not well defined: it remains unclear what interpretability means without first explicitly specifying objectives or desired attributes. In this paper, we identify the key properties used to interpret automata and propose a modification of a state-merging approach to learn variants of finite state automata. We apply the approach to problems beyond typical grammar inference tasks. Additionally, we cover several use-cases for prediction, classification, and clustering on sequential data in both supervised and unsupervised scenarios to show how the identified key properties are applicable in a wide range of contexts.



from cs.AI updates on arXiv.org http://ift.tt/2f4KEuA
via IFTTT

CAS-CNN: A Deep Convolutional Neural Network for Image Compression Artifact Suppression. (arXiv:1611.07233v1 [cs.CV])

Lossy image compression algorithms are pervasively used to reduce the size of images transmitted over the web and recorded on data storage media. However, we pay for their high compression rate with visual artifacts degrading the user experience. Deep convolutional neural networks have become a widespread tool to address high-level computer vision tasks very successfully. Recently, they have found their way into the areas of low-level computer vision and image processing to solve regression problems mostly with relatively shallow networks.

We present a novel 12-layer deep convolutional network for image compression artifact suppression with hierarchical skip connections and a multi-scale loss function. We achieve a boost of up to 1.79 dB in PSNR over ordinary JPEG and an improvement of up to 0.36 dB over the best previous ConvNet result. We show that a network trained for a specific quality factor (QF) is resilient to the QF used to compress the input image - a single network trained for QF 60 provides a PSNR gain of more than 1.5 dB over the wide QF range from 40 to 76.



from cs.AI updates on arXiv.org http://ift.tt/2fDoRWW
via IFTTT

Limbo: A Fast and Flexible Library for Bayesian Optimization. (arXiv:1611.07343v1 [cs.LG])

Limbo is an open-source C++11 library for Bayesian optimization which is designed to be both highly flexible and very fast. It can be used to optimize functions for which the gradient is unknown, evaluations are expensive, and runtime cost matters (e.g., on embedded systems or robots). Benchmarks on standard functions show that Limbo is about 2 times faster than BayesOpt (another C++ library) for a similar accuracy.



from cs.AI updates on arXiv.org http://ift.tt/2fDrkAZ
via IFTTT

Randomized Mechanisms for Selling Reserved Instances in Cloud. (arXiv:1611.07379v1 [cs.GT])

Selling reserved instances (or virtual machines) is a basic service in cloud computing. In this paper, we consider a more flexible pricing model for instance reservation, in which a customer can propose the time length and number of resources of her request, while in today's industry, customers can only choose from several predefined reservation packages. Under this model, we design randomized mechanisms for customers coming online to optimize social welfare and providers' revenue.

We first consider a simple case, where the requests from the customers do not vary too much in terms of both length and value density. We design a randomized mechanism that achieves a competitive ratio $\frac{1}{42}$ for both \emph{social welfare} and \emph{revenue}, which is a improvement as there is usually no revenue guarantee in previous works such as \cite{azar2015ec,wang2015selling}. This ratio can be improved up to $\frac{1}{11}$ when we impose a realistic constraint on the maximum number of resources used by each request. On the hardness side, we show an upper bound $\frac{1}{3}$ on competitive ratio for any randomized mechanism. We then extend our mechanism to the general case and achieve a competitive ratio $\frac{1}{42\log k\log T}$ for both social welfare and revenue, where $T$ is the ratio of the maximum request length to the minimum request length and $k$ is the ratio of the maximum request value density to the minimum request value density. This result outperforms the previous upper bound $\frac{1}{CkT}$ for deterministic mechanisms \cite{wang2015selling}. We also prove an upper bound $\frac{2}{\log 8kT}$ for any randomized mechanism. All the mechanisms we provide are in a greedy style. They are truthful and easy to be integrated into practical cloud systems.



from cs.AI updates on arXiv.org http://ift.tt/2fnfrhr
via IFTTT

Deep Learning Approximation for Stochastic Control Problems. (arXiv:1611.07422v1 [cs.LG])

Many real world stochastic control problems suffer from the "curse of dimensionality". To overcome this difficulty, we develop a deep learning approach that directly solves high-dimensional stochastic control problems based on Monte-Carlo sampling. We approximate the time-dependent controls as feedforward neural networks and stack these networks together through model dynamics. The objective function for the control problem plays the role of the loss function for the deep neural network. We test this approach using examples from the areas of optimal trading and energy storage. Our results suggest that the algorithm presented here achieves satisfactory accuracy and at the same time, can handle rather high dimensional problems.



from cs.AI updates on arXiv.org http://ift.tt/2f4IZFd
via IFTTT

An unexpected unity among methods for interpreting model predictions. (arXiv:1611.07478v1 [cs.AI])

Understanding why a model made a certain prediction is crucial in many data science fields. Interpretable predictions engender appropriate trust and provide insight into how the model may be improved. However, with large modern datasets the best accuracy is often achieved by complex models even experts struggle to interpret, which creates a tension between accuracy and interpretability. Recently, several methods have been proposed for interpreting predictions from complex models by estimating the importance of input features. Here, we present how a model-agnostic additive representation of the importance of input features unifies current methods. This representation is optimal, in the sense that it is the only set of additive values that satisfies important properties. We show how we can leverage these properties to create novel visual explanations of model predictions. The thread of unity that this representation weaves through the literature indicates that there are common principles to be learned about the interpretation of model predictions that apply in many scenarios.



from cs.AI updates on arXiv.org http://ift.tt/2fneUw3
via IFTTT

Variational Intrinsic Control. (arXiv:1611.07507v1 [cs.LG])

In this paper we introduce a new unsupervised reinforcement learning method for discovering the set of intrinsic options available to an agent. This set is learned by maximizing the number of different states an agent can reliably reach, as measured by the mutual information between the set of options and option termination states. To this end, we instantiate two policy gradient based algorithms, one that creates an explicit embedding space of options and one that represents options implicitly. The algorithms also provide an explicit measure of empowerment in a given state that can be used by an empowerment maximizing agent. The algorithm scales well with function approximation and we demonstrate the applicability of the algorithm on a range of tasks.



from cs.AI updates on arXiv.org http://ift.tt/2f4Jr6C
via IFTTT

Associative Adversarial Networks. (arXiv:1611.06953v1 [cs.LG])

We propose a higher-level associative memory for learning adversarial networks. Generative adversarial network (GAN) framework has a discriminator and a generator network. The generator (G) maps white noise (z) to data samples while the discriminator (D) maps data samples to a single scalar. To do so, G learns how to map from high-level representation space to data space, and D learns to do the opposite. We argue that higher-level representation spaces need not necessarily follow a uniform probability distribution. In this work, we use Restricted Boltzmann Machines (RBMs) as a higher-level associative memory and learn the probability distribution for the high-level features generated by D. The associative memory samples its underlying probability distribution and G learns how to map these samples to data space. The proposed associative adversarial networks (AANs) are generative models in the higher-levels of the learning, and use adversarial non-stochastic models D and G for learning the mapping between data and higher-level representation spaces. Experiments show the potential of the proposed networks.



from cs.AI updates on arXiv.org http://ift.tt/2flLnpO
via IFTTT

Alcholics anonymous

Alcholics anonymous. Monday, November 28, 2016 -. 6:30pm to 7:30pm. Brentwood Branch Library. have a meeting..help women in the community.

from Google Alert - anonymous http://ift.tt/2f4o2Kt
via IFTTT

Anonymous Just Took Down Website of Company that Sells Concussion Grenades to DAPL Cops

Anonymous just took down the website of the company who manufactures the tear gas and concussion grenades used against Standing Rock water ...

from Google Alert - anonymous http://ift.tt/2fPcXw2
via IFTTT

Anonymous Report of PCard Fraud or Abuse

Anonymous Report of PCard Fraud or Abuse. Person ALLEGED to be Fraud/Abuser*. Building & Room Number*. Department*. Date of Abuse*.

from Google Alert - anonymous http://ift.tt/2fP85qU
via IFTTT

Custom Essay Writers Anonymous

CUstom writing service. Enjoy your academic papers written from scratch by professional and experienced writers - fast, anonymous and in premium ...

from Google Alert - anonymous http://ift.tt/2fo46kg
via IFTTT

Orioles name Roger McDowell pitching coach; spent the last 11 seasons in same role with the Braves (ESPN)

from ESPN http://ift.tt/1eW1vUH
via IFTTT

[FD] [CORE-2016-0007] - TP-LINK TDDP Multiple Vulnerabilities

Ravens (5-5) drop 1 spot to No. 14 in Week 12 NFL Power Rankings next game Sunday vs. Bengals (3-6-1) (ESPN)

from ESPN http://ift.tt/17lH5T2
via IFTTT

ISS Daily Summary Report – 11/21/2016

Urine Processing Assembly (UPA) Potential Wastewater Storage Tank Assembly (WSTA) Leak: This morning an unexpected caution indicating “UPA Potential WSTA leak” annunciated following use of the Waste and Hygiene Compartment (WHC). The crew isolated the UPA and WHC was configured to internal [EDV] to isolate the system. An on-orbit inspection by the crew showed pretreated urine on the urine valve block. The crew cleaned up the pretreated urine and closed out the WHC. WHC is No-Go for use and ground specialists are working the follow-on plan.  The crew has been authorized to use the Russian toilet.  Due to the leak, the planned WSTA leak was aborted. Cygnus Departure: Cygnus was unberthed nominally from the ISS at 5:23 AM CST and was released at 7:22 AM CST. Post departure science objectives for Cygnus including Saffire-II and NanoRacks CubeSat deployments.  Cygnus re-entry is planned for November 27. Later today, Saffire-II will initiate and conduct a roughly 2-hour experiment session, followed by multi-day data downlinks. The Saffire-II experiment attempts to quantify the flammability of several materials in microgravity, and compares them to flammability limits in normal gravity.  The NanoRacks CubeSat Deployer – External (NRCSD-E) deploy is planned November 25, at which time the Cygnus will be 100 km above the ISS, this will be the the first time NRCSDs deploy above the ISS. A total of 4 LEMUR-2 satellites will deploy from the Cygnus vehicle.  The NanoRacks-LEMUR-2 satellites are part of a remote sensing satellite constellation that proves global ship tracking and weather monitoring. Marrow Collections: The crew performed their first blood collections in support of the Canadian Space Agency’s (CSA’s) Marrow experiment.  After collecting, the blood was spun in the Refrigerated Centrifuge and stored in the Minus Eighty Degree Celsius Laboratory Freezer for ISS (MELFI).   The Marrow investigation looks at the effect of microgravity on the bone marrow. It is believed that microgravity, like long-duration bed rest on Earth, has a negative effect on the bone marrow and the blood cells that are produced in the bone marrow.  Fine Motor Skills (FMS): The 49Soyuz USOS crew performed their first FMS sessions this morning.  FMS is performed regularly during the crews’ on-orbit stay, with several sessions within the first week, then as the mission progresses, more time elapses between sessions.  Each FMS session involves the crew performing a series of interactive tasks on a touchscreen tablet. This investigation is critical during long-duration space missions, particularly those skills needed to interact with technologies required in next-generation space vehicles, spacesuits, and habitats. The crewmember’s fine motor skills are also necessary for performing tasks in transit or on a planetary surface, such as information access, just-in-time training, subsystem maintenance, and medical treatment. MATISS Installation: The crew installed four MATISS sample holders in the Columbus module. The MATISS experiment investigates the antibacterial properties of materials in space to see if future spacecraft could be made easier to clean. The experiment aims to understand the mechanisms of attachment of biofilms in microgravity conditions. The optimization of the internal surfaces of the International Space Station (ISS) is a challenge. The first objective is to simplify MATISS decontamination operations to save time crew. The second objective relates to space exploration: the validation of these innovative surfaces would indeed have new assets for the development of future spacecraft, including in the context of long journeys.  Microgravity Science Laboratory (MSL) Sample Cartridge Assembly (SCA) Installation: The crew inserted a new cartridge in the MSL, located in the Material Science Research Rack (MSRR), today.  The sample processing date is scheduled for later in the increment.  The Batch-2b of the Materials Science Laboratory Sample Cartridge Assemblies (MSL SCA-Batch 2b-ESA) serves two projects investigating how different phases organize in a structure when metallic alloys are solidified. The METCOMP project studies the phase formed by the reaction of the remaining liquid phase with an already formed solid, to form a second solid phase on cooling. For this purpose, Bronze (Copper-Tin Alloys) of different compositions will be processed. The other project, Solidification along a Eutectic path in Ternary Alloys (SETA), looks at how two phases that form together organize into lamellar, or fibre, structures when cooling Aluminium (Copper-Silver Alloys). Both projects will provide benchmark samples that will enable to test numerical models that aim to predict these structures.  Mobile Servicing System (MSS) Operations:  Following the successful Cygnus unberth and release, the Robotics Ground Controllers performed a video survey of the Node 1 Nadir Common Berthing Mechanism (CBM).  They then walked the Space Station Remote Manipulator System (SSRMS) off Node 2 onto Mobile Base System (MBS) Power Data Grapple Fixture 3 (PDGF3) and maneuvered the SSRMS to a Mobile Transporter (MT) translation configuration.  Currently the Robotics Ground Controllers are performing additional troubleshooting of the SSRMS End B Camera Light Pan/Tilt Unit (PTU) Assembly (CLPA). Today’s Planned Activities All activities were completed unless otherwise noted. USOS Window Shutter Close МО-8. Setup Body Mass Measurement – BMMD CSA Generic Frozen Blood Collection Photo/TV Depress Video Setup Cygnus-Node1 Vestibule Depressurization: Part 2 МО-8. Closeout Ops CSA Generic Refrigerated Centrifuge Configure Cygnus-Node1 Vestibule Depressurization: Multimeter Reset KORREKTSIYA. Logging Liquid and Food (Medicine) Intake COSMOCARD. Photography of the Experiment Session [Aborted] CSA Generic Sample MELFI Insertion Auxiliary Laptop Computer System Virus Definition File Update CARDIOVECTOR. Experiment Ops. Fine Motor Skills Experiment Test CARDIOVECTOR. Photography of the Experiment Session Closing Shutters on windows 6, 8, 9, 12, 13, 14 CSA Generic Frozen Blood Collection Conclude And Stow Cygnus-Node1 Vestibule Depressurization: Part 3 Measuring CO partial pressure at Central Post using CSA-CP analyzer Robotic Workstation (RWS) High Definition (HD) Monitor Downlink Setup Bringing ODF up to date using Soyuz 733 delivered files Cygnus PCS Command and PROX Link Verification Cygnus-Node1 Vestibule Depressurization: Part4 Glacier Tray Insert Repositioning ПТАБ-1М (Storage Battery Current Converter) (from pos. А306 to pos.А308) and ПТАБ-1М (from pos.А308 to pos.А306), preparation Scheduled monthly maintenance of Central Post Laptop Log-File Downlink from the Laptop. ISS Crew Orientation PILOT-T. Preparation for the experiment. Cygnus-Node1 Vestibule Depressurization: Part 5 PILOT-T. Experiment Ops. Photo/TV Depress […]

from ISS On-Orbit Status Report http://ift.tt/2fBCnuo
via IFTTT

how can i give anonymous user as a group some permission in odoo?

I move from Drupal to Odoo and it is quite new to me. I think there should be specific group for anonymous users. so I can give permissions to ...

from Google Alert - anonymous http://ift.tt/2ghzgJ1
via IFTTT

[FD] [CVE-2016-7434] ntpd remote pre-auth DoS

-----BEGIN PGP SIGNED MESSAGE-

Source: Gmail -> IFTTT-> Blogger

[FD] [ERPSCAN-16-034] SAP NetWeaver AS JAVA - XXE vulnerability in BC-BMT-BPM-DSK component

[FD] [ERPSCAN-16-033] SAP NetWeaver AS JAVA icman - DoS vulnerability

Application: SAP NetWeaver AS JAVA Versions Affected: SAP NetWeaver AS JAVA 7.4 Vendor URL: http://SAP.com Bug: Denial of Service Sent: 22.04.2016 Reported: 23.04.2016 Vendor response: 23.04.2016 Date of Public Advisory: 09.08.2016 Reference: SAP Security Note 2313835 Author: Vahagn Vardanyan (ERPScan) Description 1. ADVISORY INFORMATION Title: [ERPSCAN-16-033] SAP NetWeaver AS JAVA icman – DoS vulnerability Advisory ID:[ERPSCAN-16-033] Risk: high Advisory URL: http://ift.tt/2fymFjr Date published: 11.11.2016 Vendors contacted: SAP 2. VULNERABILITY INFORMATION Class: Denial of Service Impact: Denial of Service Remotely Exploitable: yes Locally Exploitable: yes CVSS Information CVSS Base Score v3: 7.5 / 10 CVSS Base Vector: AV : Attack Vector (Related exploit range) Network (N) AC : Attack Complexity (Required attack complexity) Low (L) PR : Privileges Required (Level of privileges needed to exploit) None (N) UI : User Interaction (Required user participation) None (N) S : Scope (Change in scope due to impact caused to components beyond the vulnerable component) Unchanged (U) C : Impact to Confidentiality None (N) I : Impact to Integrity None (N) A : Impact to Availability High (H) 3. VULNERABILITY DESCRIPTION Unauthenticated attacker can make DoS attack with use P4 over HTTPS 4. VULNERABLE PACKAGES SAP KERNEL 7.21 32-BIT SAP KERNEL 7.21 32-BIT UNICODE SAP KERNEL 7.21 64-BIT SAP KERNEL 7.21 64-BIT UNICODE SAP KERNEL 7.21 EXT 32-BIT SAP KERNEL 7.21 EXT 32-BIT UC SAP KERNEL 7.21 EXT 64-BIT SAP KERNEL 7.21 EXT 64-BIT UC SAP KERNEL 7.22 64-BIT SAP KERNEL 7.22 64-BIT UNICODE SAP KERNEL 7.22 EXT 64-BIT SAP KERNEL 7.22 EXT 64-BIT UC SAP KERNEL 7.42 64-BIT SAP KERNEL 7.42 64-BIT UNICODE SAP KERNEL 7.45 64-BIT SAP KERNEL 7.45 64-BIT UNICODE 5. SOLUTIONS AND WORKAROUNDS To correct this vulnerability, install SAP Security Note 2313835 6. AUTHOR Vahagn Vardanyan (ERPScan) 7. TECHNICAL DESCRIPTION Vulnerability triggers when one sends HTTPS GET request to SAP NetWeaver P4. PoC ``` GET https://SAP_IP:50005/sap.com~P4TunnelingApp!web/myServlet HTTP/1.1 Host: 172.16.10.65:50005 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:33.0) Gecko/20100101 Firefox/33.0 Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 Accept-Language: en-US,en;q=0.5 Accept-Encoding: gzip, deflate Connection: close ``` ``` 0:007> r rax=0000a323260f1252 rbx=0000000025c500d0 rcx=0000000025c500d0 rdx=0000000000000001 rsi=0000000000000002 rdi=0000000000000000 rip=000000013f3af019 rsp=0000000003500d40 rbp=0000000003500e40 r8=0000000025c50400 r9=0000006c004c0002 r10=0000000003500c20 r11=00000000021b2df0 r12=0000000000000002 r13=000000013f2c0000 r14=0000000000000000 r15=0000000000000001 iopl=0 nv up ei ng nz ac po cy cs=0033 ss=002b ds=002b es=002b fs=0053 gs=002b efl=00010297 icman!P4PlugInReadHandler+0xb9: 00000001`3f3af019 8b4f04 mov ecx,dword ptr [rdi+4] ds:00000000`00000004=???????? 00000000`03500d40 00000001`3f363fb5 icman!P4PlugInReadHandler+0xb9 [d:\depot\bas\742_rel\src\krn\si\ic\p4_plg.c @ 1192] 00000000`03500ec0 00000001`3f3638ea icman!IcmMplxAsyncReadDone+0x75 [d:\depot\bas\742_rel\src\krn\si\ic\icxxmplx.c @ 5088] 00000000`03500f10 00000001`3f362626 icman!IcmMplxExecCall+0x36a [d:\depot\bas\742_rel\src\krn\si\ic\icxxmplx.c @ 4808] 00000000`0350fd20 00000000`74901d9f icman!IcmMplxThread+0x5f6 [d:\depot\bas\742_rel\src\krn\si\ic\icxxmplx.c @ 3840] 00000000`0350fdb0 00000000`74901e3b MSVCR100!endthreadex+0x43 00000000`0350fde0 00000000`7716652d MSVCR100!endthreadex+0xdf 00000000`0350fe10 00000000`7729c541 kernel32!BaseThreadInitThunk+0xd 00000000`0350fe40 00000000`00000000 ntdll!RtlUserThreadStart+0x21 ``` 8. REPORT TIMELINE Sent: 22.04.2016 Reported: 23.04.2016 Vendor response: 23.04.2016 Date of Public Advisory: 09.08.2016 9. REFERENCES http://ift.tt/2fymFjr 10. ABOUT ERPScan Research ERPScan research team specializes in vulnerability research and analysis of critical enterprise applications. It was acknowledged multiple times by the largest software vendors like SAP, Oracle, Microsoft, IBM, VMware, HP for discovering more than 400 vulnerabilities in their solutions (200 of them just in SAP!). ERPScan researchers are proud of discovering new types of vulnerabilities (TOP 10 Web Hacking Techniques 2012) and of the "The Best Server-Side Bug" nomination at BlackHat 2013. ERPScan experts participated as speakers, presenters, and trainers at 60+ prime international security conferences in 25+ countries across the continents ( e.g. BlackHat, RSA, HITB) and conducted private trainings for several Fortune 2000 companies. ERPScan researchers carry out the EAS-SEC project that is focused on enterprise application security awareness by issuing annual SAP security researches. ERPScan experts were interviewed in specialized infosec resources and featured in major media worldwide. Among them, there are Reuters, Yahoo, SC Magazine, The Register, CIO, PC World, DarkReading, Heise, Chinabyte, etc. Our team consists of highly-qualified researchers, specialized in various fields of cybersecurity (from web application to ICS/SCADA systems), gathering their experience to conduct the best SAP security research. 11. ABOUT ERPScan ERPScan is the most respected and credible Business Application Cybersecurity provider. Founded in 2010, the company operates globally and enables large Oil and Gas, Financial, Retail and other organizations to secure their mission-critical processes. Named as an ‘Emerging Vendor’ in Security by CRN, listed among “TOP 100 SAP Solution providers” and distinguished by 30+ other awards, ERPScan is the leading SAP SE partner in discovering and resolving security vulnerabilities. ERPScan consultants work with SAP SE in Walldorf to assist in improving the security of their latest solutions. ERPScan’s primary mission is to close the gap between technical and business security and provide solutions for CISO's to evaluate and secure SAP and Oracle ERP systems and business-critical applications from both cyberattacks and internal fraud. As a rule, our clients are large enterprises, Fortune 2000 companies and MSPs, whose requirements are to actively monitor and manage security of vast SAP and Oracle landscapes on a global scale. We ‘follow the sun’ and have two hubs, located in Palo Alto and Amsterdam, to provide threat intelligence services, continuous support and to operate local offices and partner network spanning 20+ countries around the globe. Adress USA: 228 Hamilton Avenue, Fl. 3, Palo Alto, CA. 94301 Phone: 650.798.5255 Twitter: @erpscan Scoop-it: Business Application Security

Source: Gmail -> IFTTT-> Blogger

[FD] [x33fcon] Call for Papers (and Trainers)

Howdy all! We have just opened call for papers and trainers for new security event, called x33fcon (pronounced /ˈziːf-kɒn/) taking place in Poland between April 24-28, 2017. Yet another security con? Nope. The idea behind x33fcon is to focus on a very specific need of the security industry: collaboration between blue and red teams (sometimes referred to as purple teaming). We know from our professional experience how hard it can be to build a solid bridge between those who assess security and those who guard the castle. Many misconceptions and misunderstandings arise between these two groups. We want to change that. Rather than continue to think about security from either an offensive or defensive perspective, our goal is to encourage security professionals to consider both perspectives while working more closely together. Both blue and red teams provide unique capabilities to an organization, but the collaboration between the two yields the best results. x33fcon is an event that brings both groups together, and everyone from both sides of the fence are invited to gather, share ideas and discuss thoughts on security. Topics of interest (talks): - red/blue collaboration (purple teaming) - War stories / case studies - offensive TTPs (Tactics, Techniques and Procedures) - research on novel attacks - threat intelligence - active defense - blue team best practices - networking incident response, investigation, and evidence handling - network forensics and traffic analysis - malware research - digital forensic triage - antiforensics and anti-antiforensics approaches Topics of interest (trainings): - offensive TTPs (Tactics, Techniques and Procedures) - exploit design and implementation - social engineering - malware analysis - forensics, including mobile - hardware hacking - cloud and IoT security Key deadlines: - CfP end date: Dec 31, 2016 - CfT end date: Dec 31, 2016 - Final Notification: Jan 15, 2017 - Trainings: Apr 24-26, 2017 - Conference: Apr 27-28, 2017 If you want to become a speaker or trainer at x33fcon reach out to us. More information can be found on our website: - CfP - http://ift.tt/2fXXbNa - CfT - http://ift.tt/2fN2ukO See you there! x33fcon organizers.

Source: Gmail -> IFTTT-> Blogger

[FD] MSIE8 MSHTML Ptls5::Ls­Find­Span­Visual­Boundaries memory corruption

Throughout November, I plan to release details on vulnerabilities I found in web-browsers which I've not released before. This is the fifteenth entry in that series. Unfortunately I won't be able to publish everything within one month at the current rate, so I may continue to publish these through December and January. The below information is available in more detail on my blog at http://ift.tt/2gxka5h. Follow me on http://twitter.com/berendjanwever for daily browser bugs. MSIE8 MSHTML Ptls5::LsFindSpanVisualBoundaries memory corruption ================================================================ (The fix and CVE number for this bug are unknown) Synopsis

Source: Gmail -> IFTTT-> Blogger

[FD] PHDays VII Call for Papers: How to Stand Up at the Standoff

November 15 marks the start of the call for papers for Positive Hack Days VII, an international digital security forum that will be held at the World Trade Center in Moscow, Russia, on May 23-24, 2017. Potential presenters must apply by sending an application form (http://ift.tt/2fmvgIs) before January 30, 2017. Both established experts and up-and-coming specialists are encouraged to apply. At the event, we plan to emphasize the threats and opportunities for building a safer Internet of Things. The theme of the prior forum-The Standoff-has taken on a life of its own with some unexpected twists. Cyberwarfare now threatens the everyday objects around us and hackers can use smart cars, smart home sensors, video cameras, and even children's toys to their advantage. More than just a place for discussing security issues, PHDays is a space for building solutions. We urge speakers to share specific advice and tactics for mitigating security threats. We welcome both real-world experience and original research in all imaginable areas of security: defense, offense, smart devices, social networks, physical security, social engineering... If you have something to say, we hope to hear it at PHDays VII. For information on topics and guidelines, see the Call for Papers page at http://ift.tt/2gfLq5C. Applicants are strongly encouraged to apply early. Early-bird applicants leave time for the program committee to provide recommendations, which increases the likelihood of their presentation making the final cut.

Source: Gmail -> IFTTT-> Blogger

[FD] Reflected XSS in WonderCMS <= v0.9.8

============================================= MGC ALERT 2016-006 - Original release date: Nov 16, 2016 - Last revised: Nov 21, 2016 - Discovered by: Manuel Garcia Cardenas - Severity: 4,8/10 (CVSS Base Score) ============================================= I. VULNERABILITY

Source: Gmail -> IFTTT-> Blogger

Hackers Steal Millions From European ATMs Using Malware That Spit Out Cash

ATM hackers who long relied on tactics of stealing payment card numbers and online banking credentials to steal millions are now targeting the bank itself to steal cash directly from the machines. Earlier this year, a gang of cyber criminals infected several ATMs with malware in Taiwan and Thailand that caused the machines to spit out millions in cash, and the gang members then stood in front


from The Hacker News http://ift.tt/2fmu0oM
via IFTTT

Oracle acquires DNS provider Dyn for more than $600 Million

Yes, Oracle just bought the DNS provider company that brought down the Internet last month. Business software vendor Oracle announced on Monday that it is buying cloud-based Internet performance and Domain Name System (DNS) provider Dyn. Dyn is the same company that was hit by a massive distributed denial of service (DDoS) attack by the Mirai botnet last month which knocked the entire


from The Hacker News http://ift.tt/2gx7gUP
via IFTTT

Your prepaid card can no longer be anonymous

As from end-November, anonymous top-up cards will finally be a thing of the past. You might not know this, but when you or one of your parents ...

from Google Alert - anonymous http://ift.tt/2gx5aUT
via IFTTT

Monday, November 21, 2016

Nassau County Schools Respond To Anonymous Bomb Threat

Nassau County Schools Respond To Anonymous Bomb Threat. CBS New York. Duration: 02:26 1 hr ago. SHARE · SHARE · TWEET · SHARE · EMAIL.

from Google Alert - anonymous http://ift.tt/2f14CGD
via IFTTT

Advice- Anonymous and Free

Could you use some anonymous advice on any issue or problem? I am skilled at giving practical advice on many topics. If it helps, great. If it doesn't, ...

from Google Alert - anonymous http://ift.tt/2fMvMAe
via IFTTT

Nassau County Schools Respond To Anonymous Bomb Threat

Nassau County Schools Respond To Anonymous Bomb Threat. CBS New York. Duration: 02:26 27 mins ago. SHARE · SHARE · TWEET · SHARE.

from Google Alert - anonymous http://ift.tt/2gwzsqN
via IFTTT

Invertible Conditional GANs for image editing. (arXiv:1611.06355v1 [cs.CV])

Generative Adversarial Networks (GANs) have recently demonstrated to successfully approximate complex data distributions. A relevant extension of this model is conditional GANs (cGANs), where the introduction of external information allows to determine specific representations of the generated images. In this work, we evaluate encoders to inverse the mapping of a cGAN, i.e., mapping a real image into a latent space and a conditional representation. This allows, for example, to reconstruct and modify real images of faces conditioning on arbitrary attributes. Additionally, we evaluate the design of cGANs. The combination of an encoder with a cGAN, which we call Invertible cGAN (IcGAN), enables to re-generate real images with deterministic complex modifications.



from cs.AI updates on arXiv.org http://ift.tt/2fWM9rD
via IFTTT

A Survey of Credit Card Fraud Detection Techniques: Data and Technique Oriented Perspective. (arXiv:1611.06439v1 [cs.CR])

Credit card plays a very important rule in today's economy. It becomes an unavoidable part of household, business and global activities. Although using credit cards provides enormous benefits when used carefully and responsibly,significant credit and financial damages may be caused by fraudulent activities. Many techniques have been proposed to confront the growth in credit card fraud. However, all of these techniques have the same goal of avoiding the credit card fraud; each one has its own drawbacks, advantages and characteristics. In this paper, after investigating difficulties of credit card fraud detection, we seek to review the state of the art in credit card fraud detection techniques, data sets and evaluation criteria.The advantages and disadvantages of fraud detection methods are enumerated and compared.Furthermore, a classification of mentioned techniques into two main fraud detection approaches, namely, misuses (supervised) and anomaly detection (unsupervised) is presented. Again, a classification of techniques is proposed based on capability to process the numerical and categorical data sets. Different data sets used in literature are then described and grouped into real and synthesized data and the effective and common attributes are extracted for further usage.Moreover, evaluation employed criterions in literature are collected and discussed.Consequently, open issues for credit card fraud detection are explained as guidelines for new researchers.



from cs.AI updates on arXiv.org http://ift.tt/2geNhb3
via IFTTT

Generating machine-executable plans from end-user's natural-language instructions. (arXiv:1611.06468v1 [cs.AI])

It is critical for advanced manufacturing machines to autonomously execute a task by following an end-user's natural language (NL) instructions. However, NL instructions are usually ambiguous and abstract so that the machines may misunderstand and incorrectly execute the task. To address this NL-based human-machine communication problem and enable the machines to appropriately execute tasks by following the end-user's NL instructions, we developed a Machine-Executable-Plan-Generation (exePlan) method. The exePlan method conducts task-centered semantic analysis to extract task-related information from ambiguous NL instructions. In addition, the method specifies machine execution parameters to generate a machine-executable plan by interpreting abstract NL instructions. To evaluate the exePlan method, an industrial robot Baxter was instructed by NL to perform three types of industrial tasks {'drill a hole', 'clean a spot', 'install a screw'}. The experiment results proved that the exePlan method was effective in generating machine-executable plans from the end-user's NL instructions. Such a method has the promise to endow a machine with the ability of NL-instructed task execution.



from cs.AI updates on arXiv.org http://ift.tt/2fWS4fY
via IFTTT

Fair Division via Social Comparison. (arXiv:1611.06589v1 [cs.DS])

In the classical cake cutting problem, a resource must be divided among agents with different utilities so that each agent believes they have received a fair share of the resource relative to the other agents. We introduce a variant of the problem in which we model an underlying social network on the agents with a graph, and agents only evaluate their shares relative to their neighbors' in the network. This formulation captures many situations in which it is unrealistic to assume a global view, and also exposes interesting phenomena in the original problem.

Specifically, we say an allocation is locally envy-free if no agent envies a neighbor's allocation and locally proportional if each agent values her own allocation as much as the average value of her neighbor's allocations, with the former implying the latter. While global envy-freeness implies local envy-freeness, global proportionality does not imply local proportionality, or vice versa. A general result is that for any two distinct graphs on the same set of nodes and an allocation, there exists a set of valuation functions such that the allocation is locally proportional on one but not the other.

We fully characterize the set of graphs for which an oblivious single-cutter protocol-- a protocol that uses a single agent to cut the cake into pieces --admits a bounded protocol with $O(n^2)$ query complexity for locally envy-free allocations in the Robertson-Webb model. We also consider the price of envy-freeness, which compares the total utility of an optimal allocation to the best utility of an allocation that is envy-free. We show that a lower bound of $\Omega(\sqrt{n})$ on the price of envy-freeness for global allocations in fact holds for local envy-freeness in any connected undirected graph. Thus, sparse graphs surprisingly do not provide more flexibility with respect to the quality of envy-free allocations.



from cs.AI updates on arXiv.org http://ift.tt/2fWQiM4
via IFTTT

Non-Local Color Image Denoising with Convolutional Neural Networks. (arXiv:1611.06757v1 [cs.CV])

We propose a novel deep network architecture for grayscale and color image denoising that is based on a non-local image model. Our motivation for the overall design of the proposed network stems from variational methods that exploit the inherent non-local self-similarity property of natural images. We build on this concept and introduce deep networks that perform non-local processing and at the same time they significantly benefit from discriminative learning. Experiments on the Berkeley segmentation dataset, comparing several state-of-the-art methods, show that the proposed non-local models achieve the best reported denoising performance both for grayscale and color images for all the tested noise levels. It is also worth noting that this increase in performance comes at no extra cost on the capacity of the network compared to existing alternative deep network architectures. In addition, we highlight a direct link of the proposed non-local models to convolutional neural networks. This connection is of significant importance since it allows our models to take full advantage of the latest advances on GPU computing in deep learning and makes them amenable to efficient implementations through their inherent parallelism.



from cs.AI updates on arXiv.org http://ift.tt/2fWNGO3
via IFTTT

Generalized Dropout. (arXiv:1611.06791v1 [cs.LG])

Deep Neural Networks often require good regularizers to generalize well. Dropout is one such regularizer that is widely used among Deep Learning practitioners. Recent work has shown that Dropout can also be viewed as performing Approximate Bayesian Inference over the network parameters. In this work, we generalize this notion and introduce a rich family of regularizers which we call Generalized Dropout. One set of methods in this family, called Dropout++, is a version of Dropout with trainable parameters. Classical Dropout emerges as a special case of this method. Another member of this family selects the width of neural network layers. Experiments show that these methods help in improving generalization performance over Dropout.



from cs.AI updates on arXiv.org http://ift.tt/2geLt1C
via IFTTT

Options Discovery with Budgeted Reinforcement Learning. (arXiv:1611.06824v1 [cs.LG])

We consider the problem of learning hierarchical policies for Reinforcement Learning able to discover options, an option corresponding to a sub-policy over a set of primitive actions. Different models have been proposed during the last decade that usually rely on a predefined set of options. We specifically address the problem of automatically discovering options in decision processes. We describe a new RL learning framework called Bi-POMDP, and a new learning model called Budgeted Option Neural Network (BONN) able to discover options based on a budgeted learning objective. Since Bi-POMDP are more general than POMDP, our model can also be used to discover options for classical RL tasks. The BONN model is evaluated on different classical RL problems, demonstrating both quantitative and qualitative interesting results.



from cs.AI updates on arXiv.org http://ift.tt/2fWM0V2
via IFTTT

Learning From Graph Neighborhoods Using LSTMs. (arXiv:1611.06882v1 [cs.LG])

Many prediction problems can be phrased as inferences over local neighborhoods of graphs. The graph represents the interaction between entities, and the neighborhood of each entity contains information that allows the inferences or predictions. We present an approach for applying machine learning directly to such graph neighborhoods, yielding predicitons for graph nodes on the basis of the structure of their local neighborhood and the features of the nodes in it. Our approach allows predictions to be learned directly from examples, bypassing the step of creating and tuning an inference model or summarizing the neighborhoods via a fixed set of hand-crafted features. The approach is based on a multi-level architecture built from Long Short-Term Memory neural nets (LSTMs); the LSTMs learn how to summarize the neighborhood from data. We demonstrate the effectiveness of the proposed technique on a synthetic example and on real-world data related to crowdsourced grading, Bitcoin transactions, and Wikipedia edit reversions.



from cs.AI updates on arXiv.org http://ift.tt/2geMZkf
via IFTTT

Memory Lens: How Much Memory Does an Agent Use?. (arXiv:1611.06928v1 [cs.AI])

We propose a new method to study the internal memory used by reinforcement learning policies. We estimate the amount of relevant past information by estimating mutual information between behavior histories and the current action of an agent. We perform this estimation in the passive setting, that is, we do not intervene but merely observe the natural behavior of the agent. Moreover, we provide a theoretical justification for our approach by showing that it yields an implementation-independent lower bound on the minimal memory capacity of any agent that implement the observed policy. We demonstrate our approach by estimating the use of memory of DQN policies on concatenated Atari frames, demonstrating sharply different use of memory across 49 games. The study of memory as information that flows from the past to the current action opens avenues to understand and improve successful reinforcement learning algorithms.



from cs.AI updates on arXiv.org http://ift.tt/2fWV2kE
via IFTTT

Enforcing Relational Matching Dependencies with Datalog for Entity Resolution. (arXiv:1611.06951v1 [cs.DB])

Entity resolution (ER) is about identifying and merging records in a database that represent the same real-world entity. Matching dependencies (MDs) have been introduced and investigated as declarative rules that specify ER policies. An ER process induced by MDs over a dirty instance leads to multiple clean instances, in general. General "answer sets programs" have been proposed to specify the MD-based cleaning task and its results. In this work, we extend MDs to "relational MDs", which capture more application semantics, and identify classes of relational MDs for which the general ASP can be automatically rewritten into a stratified Datalog program, with the single clean instance as its standard model.



from cs.AI updates on arXiv.org http://ift.tt/2fWRteo
via IFTTT

Coherent Dialogue with Attention-based Language Models. (arXiv:1611.06997v1 [cs.CL])

We model coherent conversation continuation via RNN-based dialogue models equipped with a dynamic attention mechanism. Our attention-RNN language model dynamically increases the scope of attention on the history as the conversation continues, as opposed to standard attention (or alignment) models with a fixed input scope in a sequence-to-sequence model. This allows each generated word to be associated with the most relevant words in its corresponding conversation history. We evaluate the model on two popular dialogue datasets, the open-domain MovieTriples dataset and the closed-domain Ubuntu Troubleshoot dataset, and achieve significant improvements over the state-of-the-art and baselines on several metrics, including complementary diversity-based metrics, human evaluation, and qualitative visualizations. We also show that a vanilla RNN with dynamic attention outperforms more complex memory models (e.g., LSTM and GRU) by allowing for flexible, long-distance memory. We promote further coherence via topic modeling-based reranking.



from cs.AI updates on arXiv.org http://ift.tt/2fWN637
via IFTTT

Context Encoders: Feature Learning by Inpainting. (arXiv:1604.07379v2 [cs.CV] UPDATED)

We present an unsupervised visual feature learning algorithm driven by context-based pixel prediction. By analogy with auto-encoders, we propose Context Encoders -- a convolutional neural network trained to generate the contents of an arbitrary image region conditioned on its surroundings. In order to succeed at this task, context encoders need to both understand the content of the entire image, as well as produce a plausible hypothesis for the missing part(s). When training context encoders, we have experimented with both a standard pixel-wise reconstruction loss, as well as a reconstruction plus an adversarial loss. The latter produces much sharper results because it can better handle multiple modes in the output. We found that a context encoder learns a representation that captures not just appearance but also the semantics of visual structures. We quantitatively demonstrate the effectiveness of our learned features for CNN pre-training on classification, detection, and segmentation tasks. Furthermore, context encoders can be used for semantic inpainting tasks, either stand-alone or as initialization for non-parametric methods.



from cs.AI updates on arXiv.org http://ift.tt/1SGbSlQ
via IFTTT

"Knowing value" logic as a normal modal logic. (arXiv:1604.08709v3 [cs.AI] UPDATED)

Recent years witness a growing interest in nonstandard epistemic logics of "knowing whether", "knowing what", "knowing how", and so on. These logics are usually not normal, i.e., the standard axioms and reasoning rules for modal logic may be invalid. In this paper, we show that the conditional "knowing value" logic proposed by Wang and Fan \cite{WF13} can be viewed as a disguised normal modal logic by treating the negation of the Kv operator as a special diamond. Under this perspective, it turns out that the original first-order Kripke semantics can be greatly simplified by introducing a ternary relation $R_i^c$ in standard Kripke models, which associates one world with two $i$-accessible worlds that do not agree on the value of constant $c$. Under intuitive constraints, the modal logic based on such Kripke models is exactly the one studied by Wang and Fan (2013,2014}. Moreover, there is a very natural binary generalization of the "knowing value" diamond, which, surprisingly, does not increase the expressive power of the logic. The resulting logic with the binary diamond has a transparent normal modal system, which sharpens our understanding of the "knowing value" logic and simplifies some previously hard problems.



from cs.AI updates on arXiv.org http://ift.tt/1W181Dx
via IFTTT

MCMC assisted by Belief Propagaion. (arXiv:1605.09042v5 [stat.ML] UPDATED)

Markov Chain Monte Carlo (MCMC) and Belief Propagation (BP) are the most popular algorithms for computational inference in Graphical Models (GM). In principle, MCMC is an exact probabilistic method which, however, often suffers from exponentially slow mixing. In contrast, BP is a deterministic method, which is typically fast, empirically very successful, however in general lacking control of accuracy over loopy graphs. In this paper, we introduce MCMC algorithms correcting the approximation error of BP, i.e., we provide a way to compensate for BP errors via a consecutive BP-aware MCMC. Our framework is based on the Loop Calculus (LC) approach which allows to express the BP error as a sum of weighted generalized loops. Although the full series is computationally intractable, it is known that a truncated series, summing up all 2-regular loops, is computable in polynomial-time for planar pair-wise binary GMs and it also provides a highly accurate approximation empirically. Motivated by this, we first propose a polynomial-time approximation MCMC scheme for the truncated series of general (non-planar) pair-wise binary models. Our main idea here is to use the Worm algorithm, known to provide fast mixing in other (related) problems, and then design an appropriate rejection scheme to sample 2-regular loops. Furthermore, we also design an efficient rejection-free MCMC scheme for approximating the full series. The main novelty underlying our design is in utilizing the concept of cycle basis, which provides an efficient decomposition of the generalized loops. In essence, the proposed MCMC schemes run on transformed GM built upon the non-trivial BP solution, and our experiments show that this synthesis of BP and MCMC outperforms both direct MCMC and bare BP schemes.



from cs.AI updates on arXiv.org http://ift.tt/27ZAKgo
via IFTTT

Word Sense Disambiguation using a Bidirectional LSTM. (arXiv:1606.03568v2 [cs.CL] UPDATED)

In this paper we present a clean, yet effective, model for word sense disambiguation. Our approach leverage a bidirectional long short-term memory network which is shared between all words. This enables the model to share statistical strength and to scale well with vocabulary size. The model is trained end-to-end, directly from the raw text to sense labels, and makes effective use of word order. We evaluate our approach on two standard datasets, using identical hyperparameter settings, which are in turn tuned on a third set of held out data. We employ no external resources (e.g. knowledge graphs, part-of-speech tagging, etc), language specific features, or hand crafted rules, but still achieve statistically equivalent results to the best state-of-the-art systems, that employ no such limitations.



from cs.AI updates on arXiv.org http://ift.tt/1PZbHEa
via IFTTT