US firm reveals gun-toting drone that can fire in mid-air

BBC News

The Tikad drone, developed by Duke Robotics, is armed with a machine-gun and a grenade launcher. However, robotics expert Professor Noel Sharkey expressed concern that gun-toting drones could make it easier to kill innocent people. For the past decade, Prof Sharkey has been campaigning against killer robots, which are fully autonomous, computer-powered weapons that would be able to track and select targets without human supervision. According to Prof Sharkey, some US military officials are concerned that although the US might follow the laws of war, terrorists could easily look at drone innovations and copy the idea to kill innocent people.

Transforming from Autonomous to Smart: Reinforcement Learning Basics


With the rapid increases in computing power, it's easy to get seduced into thinking that raw computing power can solve problems like smart edge devices (e.g., cars, trains, airplanes, wind turbines, jet engines, medical devices). In chess, the complexity of the chess piece only increases slightly (rooks can move forward and sideways a variable number of spaces, bishops can move diagonally a variable number of spaces, etc. Now think about the number and breadth of "moves" or variables that need to be considered when driving a car in a nondeterministic (random) environment: weather (precipitation, snow, ice, black ice, wind), time of day (day time, twilight, night time, sun rise, sun set), road conditions (pot holes, bumpy, slick), traffic conditions (number of vehicles, types of vehicles, different speeds, different destinations). It's nearly impossible for an autonomous car manufacturer to operate enough vehicles in enough different situations to generate the amount of data that can be virtually gathered by playing against Grand Theft Auto.

Making Predictive Models Robust: Holdout vs Cross-Validation


When evaluating machine learning models, the validation step helps you find the best parameters for your model while also preventing it from becoming overfitted. K-fold validation evaluates the data across the entire training set, but it does so by dividing the training set into K folds – or subsections – (where K is a positive integer) and then training the model K times, each time leaving a different fold out of the training data and using it instead as a validation set. Cons of the K-fold strategy: Higher computational costs; the model needs to be trained K times at the validation step (plus one more at the test step). As a former business analyst, one of his main focuses is bringing concepts of machine learning and data science to data and business analysts.

New York Yankees vs. Boston Red Sox: TV Schedule, Start Times And Series Pitching Matchups

International Business Times

The New York Yankees (60-53) host the Boston Red Sox (65-49) Friday, as the longtime rivals battle for the American League East title in a three-game series. The Red Sox, who have won eight games in a row, own a 4 1/2 game lead on the struggling Yankees. John Farrell's squad has been winning behind excellent pitching, as Boston held the Chicago White Sox and Tampa Bay Rays to a combined 13 runs over six games. Boston, who boast Cy Young favorite Chris Sale, has the best ERA (3.63) in the American League.

Computer 'anthropologists' study global fashion


Kavita Bala, professor of computer science; Noah Snavely, associate professor computer science at Cornell Tech; and Kevin Matzen, M.S. "We present a framework for visual discovery at scale, analyzing clothing and fashion across millions of images of people around the world and spanning several years," Snavely said. Bala said the group used deep learning to detect various attributes – the color or sleeve length of shirts, whether a person is wearing glasses or a hat, and so on – in millions of images. "The combination of big data, machine learning, computer vision and automated analysis algorithms makes for a very powerful analysis tool in visual discovery of fashion and other areas," Matzen said.

NASA Has Big Plans for AI on Mars and Beyond


These are two examples of how NASA hopes to use artificial intelligence. This bot-friendly future stands counter to some of the fuss in the press this past week, after Facebook shut down an experiment because two artificially intelligent bots began communicating in a shorthand language instead of English. In the future, he said, NASA astronauts could work with more intelligent robots on Mars, with the robots scouting sites and telling humans the most interesting locations to survey. "NASA is very risk-adverse [about crewed missions]," said Chien, who is technical group supervisor of the artificial intelligence group at JPL.

AI analytics expected to rise, along with management complexity


But, as AI analytics becomes more common in corporate enterprises, managing the process is expected to get more important -- and more complex. Analytics teams will have to pay more attention to "the composition of AI systems," said Donald Farmer, principal of consultancy TreeHive Strategy in Woodinville, Wash. They'll also need to implement detailed governance and oversight procedures "as companies start to put hundreds and thousands of algorithms in place," chimed in Shawn Rogers, senior director of analytic strategy at vendor Tibco Software Inc. Gartner analyst Merv Adrian foresees networks of AI-powered tools and devices that can communicate with one another and have the ability to ingest data on their own -- developments Farmer said would make it more clear that data scientists and other analysts are "participants in AI systems" as opposed to users of the technology in a traditional sense. Another issue to contend with is the level of uncertainty in what AI algorithms predict.

Artificial intelligence identifies plant species by looking at them


Machine learning algorithms have successfully identified plant species in massive herbaria just by looking at the dried specimens. "But this approach is only possible because it is based on the human expertise. It will never remove the human expertise." "Going deeper in the automated identification of Herbarium specimens" (BMC Evolutionary Biology)

Machine Learning Model Tracks U.S. Spy Planes


Marshals Service along with military aircraft and surveillance flights operated by military contractors. It turned out the sorties over the Bay Area and southern California supported of U.S. Special Operations Command training missions. The machine-learning exercise also turned up a surprising number of local and state police aerial surveillance operations in Arizona, Florida, southern California and Ohio, the web site reported. It also spotted testing of special operations aircraft based in Ohio but detected flying over other parts of the U.S.

Washington Journal Tom Simonite Discusses Future Artificial


Tom Simonite talked about the rise in artificial intelligence use at companies such as Google and Facebook and what this means for the future of workplaces across the U.S. He spoke via video link from San Francisco.