As automotive electronics continue to advance, cars are becoming more and more reliant on sensors to perform everyday driving operations. These sensors are omnipresent and help the car navigate, reduce accidents, and provide comfortable rides. However, they can also be used to learn about the drivers themselves. In this paper, we propose a method to predict, from sensor data collected at a single turn, the identity of a driver out of a given set of individuals. We cast the problem in terms of time series classification, where our dataset contains sensor readings at one turn, repeated several times by multiple drivers. We build a classifier to find unique patterns in each individual's driving style, which are visible in the data even on such a short road segment. To test our approach, we analyze a new dataset collected by AUDI AG and Audi Electronics Venture, where a fleet of test vehicles was equipped with automotive data loggers storing all sensor readings on real roads. We show that turns are particularly well-suited for detecting variations across drivers, especially when compared to straightaways. We then focus on the 12 most frequently made turns in the dataset, which include rural, urban, highway on-ramps, and more, obtaining accurate identification results and learning useful insights about driver behavior in a variety of settings.

Policy evaluation is a crucial step in many reinforcement-learning procedures, which estimates a value function that predicts states' long-term value under a given policy. In this paper, we focus on policy evaluation with linear function approximation over a fixed dataset. We first transform the empirical policy evaluation problem into a (quadratic) convex-concave saddle point problem, and then present a primal-dual batch gradient method, as well as two stochastic variance reduction methods for solving the problem. These algorithms scale linearly in both sample size and feature dimension. Moreover, they achieve linear convergence even when the saddle-point problem has only strong concavity in the dual variables but no strong convexity in the primal variables. Numerical experiments on benchmark problems demonstrate the effectiveness of our methods.

Thanks to video games, the idea of talking to a virtual character is commonplace. Whether it's the onscreen text of early Super Nintendo titles or the fully rendered characters of the Mass Effect series, people have long been able to understand how these features work. But what if onscreen characters could be used for other applications besides gaming? Kognito is a New York company focused on designing simulations and avatars. These avatars and conversational scenarios are used for applications ranging from medical professional training, helping parents learn how to teach their kids about health and substance issues and veteran care.

NEW YORK--(BUSINESS WIRE)--PulsePoint, a leading global programmatic advertising platform, today announced the next generation of its content marketing platform. Story by PulsePoint provides unified distribution across top social media channels, native and content discovery platforms and premium websites -- at global scale. This consolidates workflow and simplifies execution for content marketing. Content marketing continues to grow in importance as advertisers look for ways to differentiate themselves and connect with their core audiences in an increasingly complex media landscape. Branded content has proven successful.

Using medical imaging of 48 patients, artificial intelligence developed by researchers at the University of Adelaide was able to predict patient lifespans. The technology analysed radiological chest scans, looking for patterns to predict which of the patients would die within five years. The AI was able to predict death with a 69 percent accuracy in recently published results, the same as a trained human oncologist, and the researchers say an updated version of the technology is even more successful. We spoke with one of the study's authors, epidemiologist Lyle Palmer, to learn more. ResearchGate: What inspired this technology?

Now, Salk Institute researchers have analyzed how neurons in a critical part of the brain, called V2, respond to natural scenes, providing a better understanding of vision processing. The work is described in Nature Communications on June 8, 2017. "Understanding how the brain recognizes visual objects is important not only for the sake of vision, but also because it provides a window on how the brain works in general," says Tatyana Sharpee, an associate professor in Salk's Computational Neurobiology Laboratory and senior author of the paper. "Much of our brain is composed of a repeated computational unit, called a cortical column. In vision especially we can control inputs to the brain with exquisite precision, which makes it possible to quantitatively analyze how signals are transformed in the brain."

To be able to drive safely and reliably, autonomous cars need to have a comprehensive understanding of what's going on around them. They need to recognize other cars, trucks, motorcycles, bikes, humans, traffic lights, street signs, and everything else that may end up on or near a road. They also have to do this in all kinds of weather and lighting conditions, which is why most (if not all) companies developing autonomous cars are spending a ludicrous (but necessary) amount of time and resources collecting data in an attempt to gain experience with every possible situation. In most cases, this technique depends on humans making annotations to enormous sets of data in order to train machine learning algorithms: hundreds or thousands of people looking at snapshots or videos taken by cars driving down streets, and drawing boxes around vehicles and road signs and labeling them, over and over. Researchers from the University of Michigan think there's a better way: Doing the whole thing in simulation instead, and they've shown that it can actually be more effective than using real data annotated by humans.

While Apple recently introduced the HomePod and iOS 11, former employees who worked on Siri told the Wall Street Journal the virtual assistant is lagging behind its competitors because of company concerns about user privacy. Siri, which competes with Amazon Alexa and Google Home devices, is struggling to rise above the competition because of Apple's culture, which prioritizes user privacy, making it hard to personalize and improve the product, former Siri team employees said. Read: Amazon Sees Apple's Siri Talking With Alexa Virtual Assistant, Report Says Amazon and Google assistants had an advantage over Siri because they had more data from their search engines they could use to train their virtual assistants and have less-restrictive privacy policies than Apple, the former employees said. The Journal report said Apple protects user privacy by randomly tagging Siri searches and keeping the information tagged for a timeframe of six months, unlike its competitors Amazon and Google, which retain the data until ask for it to be discarded. That issue has delayed efforts to boost Siri because Apple "relinquished control of data before it could be used to gauge the impact of software tweaks," employees told the Journal.

If you're looking for a one-night stand, you may have a better chance of finding one now that it's almost summer. According to a new report by dating site OkCupid, 33 percent more people are looking for a one-night stand in June than during any other month of the year. The results, compiled from a survey of over 18 million members between 2013 and 2016, suggest a 17 percent increase in people's interest for a one-night stand in April, May and June, with June being the most popular month for a fling. Winter months, on the other hand, are the least ideal time for anyone looking for something casual. The same survey revealed a slight increase of 2 percent in people interested in longer term relationships during the months of January through March.

Data scientists have a variety of different skills that they bring to bear on Big Data projects. One valuable skill that is becoming popular in data science is machine learning. Machine learning is a method of data analysis that automates model building that allows computers to find hidden insights without being explicitly programmed to find a particular insight. Machine learning can be applied to data to help businesses quickly find clusters of similar objects (e.g., identify segments of customers) and to predict outcomes (e.g., identify customers who are at-risk of churning). While machine learning is a hot skill to possess, a recent study by Evans Data Corp. found that about a third of developers (36%) who are working on Big Data projects employ elements of machine learning.