LATEST LINKS
-
Harris Poll Shows 96 Percent of Americans Support Uses of Video Surveillance to Counteract Terrorism | Reuters
A recent Harris Poll survey indicates that 96 percent of U.S. citizens feel the federal government and law enforcement agencies should be able to use video surveillance in an effort to counteract terrorism and help protect U.S. citizens in specific public places... That Americans don't mind being watched is especially relevant in light of the recently exposed domestic terror plot in Boston, and subsequent FBI intelligence indicating that Al Qaida recruits are reportedly being encouraged to perform acts of terrorism inside the U.S. BRS Labs' technology blends computer vision, machine learning and artificial intelligence; it sends instant and reliable alerts to a myriad of PDA devices, and the software is compatible with all legacy camera systems. "Traditional video surveillance approaches have failed because they ignore the fact that every environment is unique," said Ray Davis, founder of BRS Labs. "These methods also require expensive, labor-intensive programming to define specific objects or activities a system should look for, so unexpected security incidents are missed," said Davis. "Any new technology approach to video surveillance must deliver the right level of protection and the right level of privacy from small, simple deployments to the most complex security environments without human intervention required." AiSighttakes visual input from a camera, learns what activities and behaviors are typical, and generates real-time alerts when it identifies activities that are not normal. It is a reasoning-based surveillance technology that functions in a manner similar to the human brain. It takes in external visual input (computer vision), while its machine learning engine observes the scene, learns and recognizes behavioral patterns and responds accordingly. Surveillance is 24/7, and since the software learns the scene, the false positives are greatly reduced.
posted by Ian Ma, 4 months - 0 comments -
Safelock - Biometric, pressure-sensitive passwords
A brilliant application of ML (neural networks). Watch the video at the bottom of the page. Won first place for most useful product at UIST 2009 Student Innovation Competition. Related article: http://news.cnet.com/8301-17938_105-10371215-1.html
posted by Ian Ma, 4 months - 0 comments -
Xavier Llorà » Blog Archive » Scaling Genetic Algorithms using MapReduce
"AS SEEN ON THE WEBSITE": "Abstract:Genetic algorithms(GAs) are increasingly being applied to large scale problems. The traditional MPI-based parallel GAs do not scale very well. MapReduce is a powerful abstraction developed by Google for making scalable and fault tolerant applications. In this paper, we mould genetic algorithms into the the MapReduce model. We describe the algorithm design and implementation of GAs on Hadoop, the open source implementation of MapReduce. Our experiments demonstrate the convergence and scalability upto 105 variable problems. Adding more resources would enable us to solve even larger problems without any changes in the algorithms and implementation."
posted by Atul, 4 months - 1 comment -
Revolutions: The difference between Statistics and Machine Learning
It's amusing reading about the cultural clash between stat and ml folks. Most people are just poking fun and having a good laugh right? Or are we shooting ourselves in the foot somehow? Either way, very entertaining picture in the article. A must-see.
posted by Ian Ma, 4 months - 0 comments -
Machine Learning by Watching and Listening
Very creative use of ML. Here are some excerpts: "Taskar, the Magerman Term Assistant Professor in the Department of Computer and Information Science, is taking machine learning to the next level. Using novel learning algorithms combining video, sound and text streams, his team has shown that computers can be taught to associate what is in a video clip with existing descriptions of characters and actions and then infer information about new material and categorize it according to what it has already learned." "Hundreds of thousands of viewers enjoy spending hours of their time writing and posting scripts of episodes on fan sites, video clips on YouTube, and information in discussion boards... From there, computers are given specialized algorithms to be able to combine the information with the video and “learn” which person is which character, what each character is doing, and with whom. At no time does anyone in the research team tag anything. This is known as “unsupervised” or “weakly” supervised learning." "Once this learning has taken place, researchers can ask the computer, “show all scenes where Kate is talking to Jack,” or “produce a montage of all scenes with swimming,” and the computer will generate the sequence. By checking on what is produced, the team then looks for patterns containing errors that suggest the algorithms and models need fine-tuning. Once the algorithm is perfected, the computer can then watch new material and add to the already known information, using its past learning to amass more knowledge."
posted by Ian Ma, 5 months - 0 comments -
Are hockey fans, scalpers ready for 'dynamic' ticket prices? - Puck Daddy - NHL - Yahoo! Sports
Dallas stars using ML to dynamically set ticket prices. + picture of cute girls
posted by Brian Donhauser, 5 months - 0 comments -
Anscombe's quartet - Wikipedia, the free encyclopedia
And interesting, simple visualization of why statistics aren't enough
posted by Sidney Burks, 6 months - 0 comments -
Some Machine Learning Libraries
Looks like a good list of ML libraries from Arturo Servin. I met him briefly on twitter and would love to get him to join us -- smart guy with a good spirit for sharing about ML. When I get the chance, I'd like to start a directory for these libraries. Useful stuff to be able to find easily.
posted by Ian Ma, 6 months - 0 comments -
Machine Learning for Beginners
A top 10 kind of thing listing machine learning resources for beginners.
posted by Ian Ma, 6 months - 0 comments