Last month, Apple CEO Tim Cook said in a televised interview that Apple was going to be around for the next "thousand years." "We are not here for a quarter or two quarters or the next quarter or the next year to next year, we are here for [a] thousand years, and so we're not about making the most, we're about making the best," Cook said. Nowhere was that more evident than at Apple's annual developers conference, which kicked off in San Francisco on Monday. The assortment of product updates and new features unveiled at the event will be available to consumers in a few months. But a close look at some of the things Apple introduced reveal a strategy that's much more far-sighted than the next iPhone release.

Last month, Apple CEO Tim Cook said in a televised interview that Apple was going to be around for the next "thousand years." "We are not here for a quarter or two quarters or the next quarter or the next year to next year, we are here for [a] thousand years, and so we're not about making the most, we're about making the best," Cook said. Nowhere was that more evident than at Apple's annual developers conference, which kicked off in San Francisco on Monday. The assortment of product updates and new features unveiled at the event will be available to consumers in a few months. But a close look at some of the things Apple introduced reveal a strategy that's much more far-sighted than the next iPhone release.

Last month, Apple CEO Tim Cook said in a televised interview that Apple was going to be around for the next "thousand years." "We are not here for a quarter or two quarters or the next quarter or the next year to next year, we are here for [a] thousand years, and so we're not about making the most, we're about making the best," Cook said. Nowhere was that more evident than at Apple's annual developers conference, which kicked off in San Francisco on Monday. The assortment of product updates and new features unveiled at the event will be available to consumers in a few months. But a close look at some of the things Apple introduced reveal a strategy that's much more far-sighted than the next iPhone release.

Google is investing significantly in artificial intelligence to power its search engines, and on Monday it announced how it's applying those investments to the enterprise sphere. At its global enterprise event series, Google Atmosphere, in Tokyo on Monday, Google executives unveiled Google Springboard, a new app that helps users find the information they need more efficiently. It also assists users by proactively providing "actionable" information and recommendations, Prabhakar Raghavan, vice president of engineering for Google Apps, said in a blog post. Raghavan noted that according to a McKinsey survey, the average knowledge worker spends the equivalent of a full day a week searching for and gathering information. Meanwhile, Google on Monday also announced a redesigned version of Google Sites, one the most popular products for enterprise customers.

The gunman who attacked a Florida LGBT nightclub had attended the club before the attack and had used a gay dating and chat app, witnesses said. Kevin West, a regular at Pulse nightclub, said Omar Mateen messaged him on and off for a year before the shooting using the gay chat and dating app Jack'd. West was dropping off a friend at the club when he noticed Mateen – whom he knew by sight but not by name – crossing the street wearing a dark cap and carrying a black cellphone about 1 a.m., an hour before the shooting. "He walked directly past me. I said, 'Hey,' and he turned and said, 'Hey,'" and nodded his head, West said.

Kevin West, a regular at Pulse, said Omar Mateen messaged him sporadically over the last year using the app Jack'd. West was dropping off a friend at the club when he noticed Mateen crossing the street wearing a dark cap and carrying a cellphone at about 1 a.m., an hour before the shooting. "He walked directly past me. I said, 'Hey,' and he turned and said, 'Hey,'" and nodded his head, West said. At least four regular customers of Pulse told the Orlando Sentinel on Monday that they believe they had seen Mateen there before.

This natural paradigm extends the Bayesian framework to dimensionality reduction tasks in higher dimensions with simpler models at greater speeds. Here an orthogonal basis is treated as a single point on a manifold and is associated with a linear subspace on which observations vary maximally. Throughout this paper, we employ the Grassmann and Stiefel manifolds for various dimensionality reduction problems, explore the connection between the two manifolds, and use Hybrid Monte Carlo for posterior sampling on the Grassmannian for the first time. We delineate in which situations either manifold should be considered. Further, matrix manifold models are used to yield scientific insight in the context of cognitive neuroscience, and we conclude that our methods are suitable for basic inference as well as accurate prediction. All datasets and computer programs are publicly available at http://www.ics.uci.edu/

The purpose if this master's thesis is to study and develop a new algorithmic framework for Collaboartive Filtering to produce recommendations in the top-N recommendation problem. Thus, we propose Lanczos Latent Factor Recommender (LLFR); a novel "big data friendly" collaborative filtering algorithm for top-N recommendation. Using a computationally efficient Lanczos-based procedure, LLFR builds a low dimensional item similarity model, that can be readily exploited to produce personalized ranking vectors over the item space. A number of experiments on real datasets (MovieLens10M, Yahoo!Music) at different density levels indicate that LLFR outperforms other state-of-the-art top-N recommendation methods from a computational as well as a qualitative perspective. Our experimental results also show that its relative performance gains, compared to competing methods, increase as the data get sparser, where there is not enough data for the system to uncover similarities and generate reliable recommendations. More specifically, this is true both when the sparsity is generalized - as in the New Community Problem, a very common problem faced by real recommender systems in their beginning stages, when there is not sufficient number of ratings for the collaborative filtering algorithms to uncover similarities between items or users - and in the very interesting case where the sparsity is localized in a small fraction of the dataset - as in the New Users Problem, where new users are introduced to the system, they have not rated many items and thus, the CF algorithm can not make reliable personalized recommendations yet.

The choice of approximate posterior distribution is one of the core problems in variational inference. Most applications of variational inference employ simple families of posterior approximations in order to allow for efficient inference, focusing on mean-field or other simple structured approximations. This restriction has a significant impact on the quality of inferences made using variational methods. We introduce a new approach for specifying flexible, arbitrarily complex and scalable approximate posterior distributions. Our approximations are distributions constructed through a normalizing flow, whereby a simple initial density is transformed into a more complex one by applying a sequence of invertible transformations until a desired level of complexity is attained. We use this view of normalizing flows to develop categories of finite and infinitesimal flows and provide a unified view of approaches for constructing rich posterior approximations. We demonstrate that the theoretical advantages of having posteriors that better match the true posterior, combined with the scalability of amortized variational approaches, provides a clear improvement in performance and applicability of variational inference.

In this paper we study the probabilistic properties of the posteriors in a speech recognition system that uses a deep neural network (DNN) for acoustic modeling. We do this by reducing Kaldi's DNN shared pdf-id posteriors to phone likelihoods, and using test set forced alignments to evaluate these using a calibration sensitive metric. Individual frame posteriors are in principle well-calibrated, because the DNN is trained using cross entropy as the objective function, which is a proper scoring rule. When entire phones are assessed, we observe that it is best to average the log likelihoods over the duration of the phone. Further scaling of the average log likelihoods by the logarithm of the duration slightly improves the calibration, and this improvement is retained when tested on independent test data.