The big problem is that reinforcement learning is a difficult technique to characterise. Put simply an RL system learns not by being told how close it is the the desired result, but by receiving rewards based on its behaviour. Of course this is largely how we learn and if it can be made to work efficiently it promises us not just effective AI but new knowledge. For example AlphaGo taught itself to play Go and in the process discovered for itself approaches to Go that humans had ignored.

The following interview is one of many included in the report. Brendan Frey is a co-founder of Deep Genomics, a professor at the University of Toronto and a co-founder of its Machine Learning Group, a senior fellow of the Neural Computation program at the Canadian Institute for Advanced Research and a fellow of the Royal Society of Canada. His work focuses on using machine learning to understand the genome and to realize new possibilities in genomic medicine. I completed my Ph.D. with Geoff Hinton in 1997. We co-authored one of the first papers on deep learning, published in Science in 1995.

Google has been given access to huge swatches of confidential patient information in the UK, raising fears yet again over how NHS managers view and handle data under their control. In an agreement uncovered by the New Scientist, Google and its DeepMind artificial intelligence wing have been granted access to current and historic patient data at three London hospitals run by the Royal Free NHS Trust, covering 1.6 million individuals. That would include any chronic illness people may be suffering from and the circumstances over why they were admitted – for example, if they have suffered a drug overdose. The agreement provides Google with access to data going back five years and is far more expansive than expected. Google and DeepMind previously said they were working with the NHS on a product called "Streams" that would "present timely information that helps nurses and doctors detect cases of acute kidney injury." The agreement however provides access to all patient data, covering issues far beyond just kidney functioning.

We present a theoretical analysis and empirical evaluations of a novel set of techniques for computational cost reduction of classifiers that are based on learned transform and soft-threshold. By modifying optimization procedures for dictionary and classifier training, as well as the resulting dictionary entries, our techniques allow to reduce the bit precision and to replace each floating-point multiplication by a single integer bit shift. We also show how the optimization algorithms in some dictionary training methods can be modified to penalize higher-energy dictionaries. We applied our techniques with the classifier Learning Algorithm for Soft-Thresholding, testing on the datasets used in its original paper. Our results indicate it is feasible to use solely sums and bit shifts of integers to classify at test time with a limited reduction of the classification accuracy. These low power operations are a valuable trade off in FPGA implementations as they increase the classification throughput while decrease both energy consumption and manufacturing cost.

In this paper we present a method for learning a discriminative classifier from unlabeled or partially labeled data. Our approach is based on an objective function that trades-off mutual information between observed examples and their predicted categorical class distribution, against robustness of the classifier to an adversarial generative model. The resulting algorithm can either be interpreted as a natural generalization of the generative adversarial networks (GAN) framework or as an extension of the regularized information maximization (RIM) framework to robust classification against an optimal adversary. We empirically evaluate our method - which we dub categorical generative adversarial networks (or CatGAN) - on synthetic data as well as on challenging image classification tasks, demonstrating the robustness of the learned classifiers. We further qualitatively assess the fidelity of samples generated by the adversarial generator that is learned alongside the discriminative classifier, and identify links between the CatGAN objective and discriminative clustering algorithms (such as RIM).

Earlier this week, OpenAI, the nonprofit research group with billion-dollar backing from Elon Musk and other tech luminaries, launched its first program. It's called OpenAI Gym, and it's meant to be used as a benchmarking tool for artificial intelligence programs. Musk once said he thought truly artificial intelligent agents could be more harmful to the human race than nuclear weapons. When OpenAI was launched in December, its stated goal was to "advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return." Which sounds a lot like an AI version of Google's long-held mantra: "Don't be evil."

For the past three years, our smartest engineers at Databricks have been working on a stealth project. Today, we are unveiling DeepSpark, a major new milestone in Apache Spark. DeepSpark uses cutting-edge neural networks to automate the many manual processes of software development, including writing test cases, fixing bugs, implementing features according to specs, and reviewing pull requests (PRs) for their correctness, simplicity, and style. Scaling Spark's development has been a top priority for us. Every year, Spark's popularity reaches new highs.

One of my favorite small projects, EmailFiler, was motivated by a school assignment for Georgia Tech's Intro to Machine Learning class. Basically, the assignment was to pick some datasets, throw a bunch of supervised learning algorithms at them, and analyze the results. But here's the thing: we could make our own datasets if we so chose. And so choose I did - to export my gmail data and explore the feasibility of machine-learned email categorization. See, I learned long ago that it's often best to keep emails around in case there is randomly some need to refer back to them in the future.

Guest blog post by Christopher Dole and other contributors, originally posted here. Deep Learning is one of the most revolutionary and disruptive technologies ever developed in Data Science. Essentially, this is a class of algorithms inspired by how the human brain works, and it has the ability to automate and replace most of the world's jobs. This is what enables self-driving cars to function and what allows Spotify to create very customized playlists and recommendations. This is how YouTube is able to identify faces and animals in videos and how Siri can understand and process free speech in milliseconds.

Deep Learning is one of the most revolutionary and disruptive technologies ever developed in Data Science. Essentially, this is a class of algorithms inspired by how the human brain works, and it has the ability to automate and replace most of the world's jobs. This is what enables self-driving cars to function and what allows Spotify to create very customized playlists and recommendations. This is how YouTube is able to identify faces and animals in videos and how Siri can understand and process free speech in milliseconds. Deep Learning has also led to several recent advancements in healthcare.