This analysis explores the temporal sequencing of objects in a movie trailer. Temporal sequencing of objects in a movie trailer (e.g., a long shot of an object vs intermittent short shots) can convey information about the type of movie, plot of the movie, role of the main characters, and the filmmakers cinematographic choices. When combined with historical customer data, sequencing analysis can be used to improve predictions of customer behavior. E.g., a customer buys tickets to a new movie and maybe the customer has seen movies in the past that contained similar sequences. To explore object sequencing in movie trailers, we propose a video convolutional network to capture actions and scenes that are predictive of customers' preferences. The model learns the specific nature of sequences for different types of objects (e.g., cars vs faces), and the role of sequences in predicting customer future behavior. We show how such a temporal-aware model outperforms simple feature pooling methods proposed in our previous works and, importantly, demonstrate the additional model explain-ability allowed by such a model.

Topic sparsity refers to the observation that individual documents usually focus on several salient topics instead of covering a wide variety of topics, and a real topic adopts a narrow range of terms instead of a wide coverage of the vocabulary. Understanding this topic sparsity is especially important for analyzing user-generated web content and social media, which are featured in the form of extremely short posts and discussions. As topic sparsity of individual documents in online social media increases, so does the difficulty of analyzing the online text sources using traditional methods. In this paper, we propose two novel neural models by providing sparse posterior distributions over topics based on the Gaussian sparsemax construction, enabling efficient training by stochastic backpropagation. We construct an inference network conditioned on the input data and infer the variational distribution with the relaxed Wasserstein (RW) divergence. Unlike existing works based on Gaussian softmax construction and Kullback-Leibler (KL) divergence, our approaches can identify latent topic sparsity with training stability, predictive performance, and topic coherence. Experiments on different genres of large text corpora have demonstrated the effectiveness of our models as they outperform both probabilistic and neural methods.

The conversation about unconscious bias in artificial intelligence often focuses on algorithms that unintentionally cause disproportionate harm to entire swaths of society--those that wrongly predict black defendants will commit future crimes, for example, or facial-recognition technologies developed mainly by using photos of white men that do a poor job of identifying women and people with darker skin. But the problem could run much deeper than that. Society should be on guard for another twist: the possibility that nefarious actors could seek to attack artificial intelligence systems by deliberately introducing bias into them, smuggled inside the data that helps those systems learn. This could introduce a worrisome new dimension to cyberattacks, disinformation campaigns or the proliferation of fake news. According to a U.S. government study on big data and privacy, biased algorithms could make it easier to mask discriminatory lending, hiring or other unsavory business practices.

Important resources like minerals, oil and diamonds often go hand-in-hand with conflict and poor governance. But when it comes to one particular resource -- the most important resource of all -- many think a different theory will hold true. Often referred to as the water wars thesis, it suggests that growing water scarcity will drive violent conflict as access to water dries up for certain communities. Analysts worry that people, opportunistic politicians and powerful corporations will battle for dwindling water supply, inflaming tensions. In a new study, researchers tried to map out how water wars will emerge around the world and which countries are most likely to see water-related conflict in the coming decades.

When it comes to artificial intelligence, the United States and China are viewed as the industry leaders.

This week's milestones in the history of technology include the coining of the term "artificial intelligence," the digitization of the Library of Congress, and the first penny paper. The first issue of Scientific American is published by Rufus Porter as a weekly broadsheet subtitled "The Advocate of Industry and Enterprise, and Journal of Mechanical and Other Improvements." In an era of rapid innovation, Scientific American founded the first branch of the U.S. Patent Agency, in 1850, to provide technical help and legal advice to inventors. A Washington, D.C., branch was added in 1859. By 1900 more than 100,000 inventions had been patented thanks to Scientific American.

Attribute-aware CF models aims at rating prediction given not only the historical rating from users to items, but also the information associated with users (e.g. age), items (e.g. price), or even ratings (e.g. rating time). This paper surveys works in the past decade developing attribute-aware CF systems, and discovered that mathematically they can be classified into four different categories. We provide the readers not only the high level mathematical interpretation of the existing works in this area but also the mathematical insight for each category of models. Finally we provide in-depth experiment results comparing the effectiveness of the major works in each category.

It's used internally and to me it's the perfect thickness of abstraction for DL research if you use TF. I write a lot of custom layers and while there are a few TF quirks you have to know, Sonnet has much less mental overhead than the TF.layers lib and is way more "hackable." I tried out all the other topper libs pretty extensively and Sonnet really stood out. The main issue with external adoption, is, well, that there is none _ . I tried looking up a DCGAN example in Sonnet and couldn't find an open source one...there are lots internally, though.

A community for discussion and news related to Natural Language Processing (NLP). Natural language processing (NLP) is a field of computer science, artificial intelligence and computational linguistics concerned with the interactions between computers and human (natural) languages, and, in particular, concerned with programming computers to fruitfully process large natural language corpora.

It was 2014, around the time when Travis Kalanick referred to Uber as his chick-magnet "Boober" in a GQ article, that I'd realized congestion in San Francisco had gone insane. Before there was Uber, getting across town took about ten minutes by car and there was nowhere to park, ever. With Boober in play, there was parking in places there never were spaces, but the streets were so jammed with empty, one-person "gig economy" cars circling, sitting in bus zones, mowing down bicyclists whilst fussing with their phones, still endlessly going nowhere, alone, that walking across the city was faster. To be fair, you wouldn't know there were 5,700 more vehicles a day on our roads if you'd just moved here. Nor if you were pouring Uber-delivered champagne over yourself in a tub of stock options while complaining about San Francisco's homeless from the comfort of your company-rental Airbnb where artists or Mexican families once lived.