We propose a principle for exploring context in machine learning models. Starting with a simple assumption that each observation (random variables) may or may not depend on its context (conditional variables), a conditional probability distribution is decomposed into two parts: context-free and context-sensitive. Then by employing the log-linear word production model for relating random variables to their embedding space representation and making use of the convexity of natural exponential function, we show that the embedding of an observation can also be decomposed into a weighted sum of two vectors, representing its context-free and context-sensitive parts, respectively. This simple treatment of context provides a unified view of many existing deep learning models, leading to revisions of these models able to achieve significant performance boost. Specifically, our upgraded version of a recent sentence embedding model (Arora et al., 2017) not only outperforms the original one by a large margin, but also leads to a new, principled approach for compositing the embeddings of bag-of-words features, as well as a new architecture for modeling attention in deep neural networks. More surprisingly, our new principle provides a novel understanding of the gates and equations defined by the long short term memory (LSTM) model, which also leads to a new model that is able to converge significantly faster and achieve much lower prediction errors. Furthermore, our principle also inspires a new type of generic neural network layer that better resembles real biological neurons than the traditional linear mapping plus nonlinear activation based architecture. Its multi-layer extension provides a new principle for deep neural networks which subsumes residual network (ResNet) as its special case, and its extension to convolutional neutral network model accounts for irrelevant input (e.g., background in an image) in addition to filtering. Our models are validated through a series of benchmark datasets and we show that in many cases, simply replacing existing layers with our context-aware counterparts is sufficient to significantly improve the results.

In this work, we introduce the concept of bandlimiting into the theory of machine learning because all physical processes are bandlimited by nature, including real-world machine learning tasks. After the bandlimiting constraint is taken into account, our theoretical analysis has shown that all practical machine learning tasks are asymptotically solvable in a perfect sense. Furthermore, the key towards this solvability almost solely relies on two factors: i) a sufficiently large amount of training samples beyond a threshold determined by a difficulty measurement of the underlying task; ii) a sufficiently complex model that is properly bandlimited. Moreover, for some special cases, we have derived new error bounds for perfect learning, which can quantify the difficulty of learning. These case-specific bounds are much tighter than the uniform bounds in conventional learning theory. Our results have provided a new perspective to explain the recent successes of large-scale supervised learning using complex models like neural networks.

Deep neural networks (DNNs) have been widely used in the fields such as natural language processing, computer vision and image recognition. But several studies have been shown that deep neural networks can be easily fooled by artificial examples with some perturbations, which are widely known as adversarial examples. Adversarial examples can be used to attack deep neural networks or to improve the robustness of deep neural networks. A common way of generating adversarial examples is to first generate some noises and then add them into original examples. In practice, different examples have different noise-sensitive. To generate an effective adversarial example, it may be necessary to add a lot of noise to low noise-sensitive example, which may make the adversarial example meaningless. In this paper, we propose a noise-sensitivity-analysis-based test prioritization technique to pick out examples by their noise sensitivity. We construct an experiment to validate our approach on four image sets and two DNN models, which shows that examples are sensitive to noise and our method can effectively pick out examples by their noise sensitivity.

Trends in terrestrial temperature variability are perhaps more relevant for species viability than trends in mean temperature. In this paper, we develop methodology for estimating such trends using multi-resolution climate data from polar orbiting weather satellites. We derive two novel algorithms for computation that are tailored for dense, gridded observations over both space and time. We evaluate our methods with a simulation that mimics these data's features and on a large, publicly available, global temperature dataset with the eventual goal of tracking trends in cloud reflectance temperature variability.

Royal Dutch Shell is heavily investing in research and development of artificial intelligence (AI), which it hopes will provide solutions to some of its most pressing challenges. From meeting the demands of a transitioning energy market, urgently in need of cleaner and more efficient power, to improving safety on the forecourts of its service stations, AI is at the top of the agenda. I have been working with Shell over the past months to help create a data strategy, which gave me a thorough insight into Shell's AI priorities and initiatives. Current initiatives include deploying reinforcement learning in its exploration and drilling program, to reduce the cost of extracting the gas that still drives a significant proportion of its revenues. Elsewhere across its global business, Shell is rolling out AI at its public electric car charging stations, to manage the shifting demand for power throughout a day.

India ranks third in the world in terms of high quality research publications in artificial intelligence (AI) but is at a significant distance from world leader China, according to an analysis by research agency Itihaasa, which was founded by Kris Gopalakrishnan, former CEO and co-founder of Infosys. The agency computed the number of'citable documents'-- the number of research publications in peer-reviewed journals -- in the field of AI between 2013-2017 as listed out by Scimago, a compendium that tracks trends in scientific research publications. India, while third in the world with 12,135 documents, trailed behind China with 37, 918 documents and the United States with 32,421 documents. However, when parsed by another metric'citations'-- or the number of times an article is referenced -- India ranked only fifth and trailed the United Kingdom, Canada, the U.S. and China. "This suggests that India must work at improving the quality of its research output in AI," said Dayasindhu N., one of the authors of the report'Landscape of AI/ML (Machine Learning) Research In India'.

Nvidia is a company that has reached the highest highs and the lowest lows, all in the span of a couple of weeks. TechCrunch is experimenting with new content forms. This is a rough draft of something new -- provide your feedback directly to the author (Danny at danny@techcrunch.com) if you like or hate something here. Over the past two months, Nvidia's stock has dropped from a closing price of $289.36 on October 1 to today's opening of $148.42, a decline of 48.8 percent. It takes a lot for a company to lose nearly half its value in such a short period of time, but Nvidia is proving that an otherwise strong technology business can disappear in the blink of an eye.

Rank Watch is a toolset for SEO professionals that provides Internet marketing tools for search engine optimization ("SEO") social media management (SMM) website optimization, including research and analysis, link building, campaign management, automated tracking of search engine performance, analytics and conversion tracking, and SEO reports. These services are provided to you through the site based on the plan purchased, including all software, data, text, images, sounds, videos, and other content made available through the site, or developed via the Rank Watch API (collectively, "Content"). Any new features added to or augmenting the Service, are also subject to these Terms. Rank Watch provides a free account and several tiered service, fee based accounts. Fees are based on the package the user has chosen.

An increasing number of countries, companies and regions are embracing sustainable energy generation and the landscape is rapidly evolving. Here are 6 renewable energy trends to watch in the coming year. Renewable energy is booming in China.Getty Energy storage plays an important role in balancing power supply and demand, and is key to tackling the intermittency issues of renewable energy. Pairing a storage system with a renewable energy source ensures a smooth and steady power supply, even when weather conditions are not optimal for energy generation. Batteries are the most common storage devices used in renewable energy systems and their use is increasing on both the residential and grid-wide scale.

A Twitter bug meant that private tweets were made public, the site has admitted. Android users who had kept their private for more than four years were vulnerable to the bug, which would have exposed their posts despite them having chosen for them not to be public. The company allows users to protect tweets, hiding them from public view so that only approved people can follow and read posts from an account. Twitter users often protect their tweets because allowing anyone to read them might endanger them or cause other problems. Those using Twitter for Android may have been affected by the bug if they made changes to their account's settings, such as changing the email address they use on their account, Twitter said.