In San Francisco, a young engineer hopes to "optimise" his life through sensors that track his heart rate, respiration and sleep cycle. In Copenhagen, a bus running two minutes behind schedule transmits its location and passenger count to the municipal traffic signal network, which extends the time of the green light at each of the next three intersections long enough for its driver to make up some time. In Davao City in the Philippines, an unsecured webcam overlooks the storeroom of a fast food stand, allowing anyone to peer in on all its comings and goings. What links these wildly different circumstances is a vision of connected devices now being sold to us as the "internet of things". The technologist Mike Kuniavsky, a pioneer of this idea, characterises it as a state of being in which "computation and data communication [are] embedded in, and distributed through, our entire environment".

LG Electronics' vice president David VanderWaal and Amazon Echo vice president Mike George present the LG Smart InstaView Door-in-Door Refrigerator to CES 2017 attendees at the LG Electronics press conference on Wednesday, Jan. 4, 2017, in Las Vegas. Amazon wants you to take wardrobe advice from a connected gadget. It brought personal digital assistants into our kitchens with Echo, a connected speaker. Now it can see, so Amazon wants to come into our bedrooms to help us choose more flattering outfits. The new Echo Look appliance features a depth-sensing camera, built-in lighting and Style Check software that uses the latest advances in machine learning.

In 2011, Apple became the first company to place artificial intelligence in the pockets of millions of consumers when it launched the voice assistant Siri on the iPhone. Six years later, the technology giant is struggling to find its voice in AI. Analysts say the question of whether Apple can succeed in building great artificial-intelligence products is as fundamental to the company's next decade as the iPhone was to its previous one. But the tech giant faces a formidable dilemma because the nature of artificial intelligence pushes Apple far out of its comfort zone of sleekly designed hardware and services. AI programming demands a level of data collection and mining that is at odds with Apple's rigorous approach to privacy, as well as its positioning as a company that doesn't profile consumers.

Apple has announced Homepod, a new voice-controlled speaker. The device will compete with Amazon's Alexa and Google Home. It will be released later this year, the company said.

Not available

Apple is touting the audio qualities of its talking speaker, but there's much more at stake.

Apple's new HomePod speaker, coming at the end of the year, has deep rich sound that blows the Amazon Echo and Google Home out of the water. Apple developers debuted their new home device which features Siri and Apple Music at the WWDC Conference in San Francisco, California. After spending time listening to several songs on the HomePod here at the Apple Worldwide Developers Conference here, I, was wowed. But that doesn't mean Apple will have an easy time of it. Sales of high-end audio have never been a consumer electronics topper.

That was the first of many questions I asked the HomePod Apple installed in the corner of its WWDC demo area, and the answer was the same each time: silence. Sure, the hazy light on top of the speaker ebbed and flowed -- that doesn't mean it understood anything I was saying. Spokespeople quickly clarified that this was a non-functional demo unit, but I did eventually hear one next to a Sonos PLAY:3 and an original Amazon Echo. Musically, it blew them both out of the water. It looks almost exactly the way countless leaks and reports suggested: It's a short, rounded cylinder covered in a fine fabric mesh with a blinky light on top.

Multi-Task Learning (MTL) can enhance a classifier's generalization performance by learning multiple related tasks simultaneously. Conventional MTL works under the offline or batch setting, and suffers from expensive training cost and poor scalability. To address such inefficiency issues, online learning techniques have been applied to solve MTL problems. However, most existing algorithms of online MTL constrain task relatedness into a presumed structure via a single weight matrix, which is a strict restriction that does not always hold in practice. In this paper, we propose a robust online MTL framework that overcomes this restriction by decomposing the weight matrix into two components: the first one captures the low-rank common structure among tasks via a nuclear norm and the second one identifies the personalized patterns of outlier tasks via a group lasso. Theoretical analysis shows the proposed algorithm can achieve a sub-linear regret with respect to the best linear model in hindsight. Even though the above framework achieves good performance, the nuclear norm that simply adds all nonzero singular values together may not be a good low-rank approximation. To improve the results, we use a log-determinant function as a non-convex rank approximation. The gradient scheme is applied to optimize log-determinant function and can obtain a closed-form solution for this refined problem. Experimental results on a number of real-world applications verify the efficacy of our method.

Recommender systems are widely used to predict personalized preferences of goods or services using users' past activities, such as item ratings or purchase histories. If collections of such personal activities were made publicly available, they could be used to personalize a diverse range of services, including targeted advertisement or recommendations. However, there would be an accompanying risk of privacy violations. The pioneering work of Narayanan et al.\ demonstrated that even if the identifiers are eliminated, the public release of user ratings can allow for the identification of users by those who have only a small amount of data on the users' past ratings. In this paper, we assume the following setting. A collector collects user ratings, then anonymizes and distributes them. A recommender constructs a recommender system based on the anonymized ratings provided by the collector. Based on this setting, we exhaustively list the models of recommender systems that use anonymized ratings. For each model, we then present an item-based collaborative filtering algorithm for making recommendations based on anonymized ratings. Our experimental results show that an item-based collaborative filtering based on anonymized ratings can perform better than collaborative filterings based on 5--10 non-anonymized ratings. This surprising result indicates that, in some settings, privacy protection does not necessarily reduce the usefulness of recommendations. From the experimental analysis of this counterintuitive result, we observed that the sparsity of the ratings can be reduced by anonymization and the variance of the prediction can be reduced if $k$, the anonymization parameter, is appropriately tuned. In this way, the predictive performance of recommendations based on anonymized ratings can be improved in some settings.