IBM's technology has won Jeopardy, managed companies and is now practicing law. ROSS, 'the world's first artificially intelligent attorney' powered by Watson, recently landed a position at New York law firm Baker & Hostetler handling the firm's bankruptcy practice. The machine is designed to understand language, provide answers to questions, formulate hypotheses and monitor developments in the legal system. IBM's technology has won Jeopardy, managed companies and is now practicing law. ROSS, 'the world's first artificially intelligent attorney' powered by Watson, has just landed a position at New York law firm Baker & Hostetler handling the firm's bankruptcy practice Lawyers ask ROSS research questions in natural language, just like they were talking to a colleague, and the AI'reads' through the law, gathers evidence, draws inferences and returns with a'highly relevant', evidence-based answer.

Exploration-exploitation of functions, that is learning and optimizing a mapping between inputs and expected outputs, is ubiquitous to many real world situations. These situations sometimes require us to avoid certain outcomes at all cost, for example because they are poisonous, harmful, or otherwise dangerous. We test participants' behavior in scenarios in which they have to find the optimum of a function while at the same time avoid outputs below a certain threshold. In two experiments, we find that Safe-Optimization, a Gaussian Process-based exploration-exploitation algorithm, describes participants' behavior well and that participants seem to care firstly whether a point is safe and then try to pick the optimal point from all such safe points. This means that their trade-off between exploration and exploitation can be seen as an intelligent, approximate, and homeostasis-driven strategy.

Databases in domains such as healthcare are routinely released to the public in aggregated form. Unfortunately, naïve modeling with aggregated data may significantly diminish the accuracy of inferences at the individual level. This paper addresses the scenario where features are provided at the individual level, but the target variables are only available as histogram aggregates or order statistics. We consider a limiting case of generalized linear modeling when the target variables are only known up to permutation, and explore how this relates to permutation testing; a standard technique for assessing statistical dependency. Based on this relationship, we propose a simple algorithm to estimate the model parameters and individual level inferences via alternating imputation and standard generalized linear model fitting. Our results suggest the effectiveness of the proposed approach when, in the original data, permutation testing accurately ascertains the veracity of the linear relationship. The framework is extended to general histogram data with larger bins - with order statistics such as the median as a limiting case. Our experimental results on simulated data and aggregated healthcare data suggest a diminishing returns property with respect to the granularity of the histogram - when a linear relationship holds in the original data, the targets can be predicted accurately given relatively coarse histograms.

Learning the true ordering between objects by aggregating a set of expert opinion rank order lists is an important and ubiquitous problem in many applications ranging from social choice theory to natural language processing and search aggregation. We study the problem of unsupervised rank aggregation where no ground truth ordering information in available, neither about the true preference ordering between any set of objects nor about the quality of individual rank lists. Aggregating the often inconsistent and poor quality rank lists in such an unsupervised manner is a highly challenging problem, and standard consensus-based methods are often ill-defined, and difficult to solve. In this manuscript we propose a novel framework to bypass these issues by using object attributes to augment the standard rank aggregation framework. We design algorithms that learn joint models on both rank lists and object features to obtain an aggregated rank ordering that is more accurate and robust, and also helps weed out rank lists of dubious validity. We validate our techniques on synthetic datasets where our algorithm is able to estimate the true rank ordering even when the rank lists are corrupted. Experiments on three real datasets, MQ2008, MQ2008 and OHSUMED, show that using object features can result in significant improvement in performance over existing rank aggregation methods that do not use object information. Furthermore, when at least some of the rank lists are of high quality, our methods are able to effectively exploit their high expertise to output an aggregated rank ordering of great accuracy.

This volume is a collection of contributions from the 5th Workshop on Machine Learning and Interpretation in Neuroimaging (MLINI) at the Neural Information Processing Systems (NIPS 2015) conference. Modern multivariate statistical methods developed in the rapidly growing field of machine learning are being increasingly applied to various problems in neuroimaging, from cognitive state detection to clinical diagnosis and prognosis. Multivariate pattern analysis methods are designed to examine complex relationships between high-dimensional signals, such as brain images, and outcomes of interest, such as the category of a stimulus, a type of a mental state of a subject, or a specific mental disorder. Such techniques are in contrast with the traditional mass-univariate approaches that dominated neuroimaging in the past and treated each individual imaging measurement in isolation. We believe that machine learning has a prominent role in shaping how questions in neuroscience are framed, and that the machine-learning mind set is now entering modern psychology and behavioral studies. It is also equally important that practical applications in these fields motivate a rapidly evolving line or research in the machine learning community. In parallel, there is an intense interest in learning more about brain function in the context of rich naturalistic environments and scenes. Efforts to go beyond highly specific paradigms that pinpoint a single function, towards schemes for measuring the interaction with natural and more varied scene are made. The goal of the workshop is to pinpoint the most pressing issues and common challenges across the neuroscience, neuroimaging, psychology and machine learning fields, and to sketch future directions and open questions in the light of novel methodology.

Rescale's Design-of-Experiments (DOE) framework is an easy way to optimize the performance of machine learning models. This article will discuss a workflow for doing hyper-parameter optimization on deep neural networks. For an introduction to DOEs on Rescale, see this webinar. Deep neural networks (DNNs) are a popular machine learning model used today in many many applications including robotics, self-driving cars, image search, facial recognition, and speech recognition. In this article we will train some neural networks to do image classification and show how to use Rescale to maximize the performance of your DNN models.

For good reason, deep learning is increasingly capturing mainstream attention. Just recently, on March 15th, Google DeepMind's AlphaGo AI -- technology based on deep neural networks -- beat Lee Sedol, one of the world's best Go players, in a professional Go match. Behind the scenes, deep learning is an active, fast-paced research area that's proliferating quickly among some of the world's most innovative companies. We are asked frequently about our favorite resources to get up to speed on deep learning and follow its rapid developments. As such, we've outlined below some of our favorite resources. While certainly not comprehensive, there's a lot here, and we'll continue to update this list -- if there's something we should add, let us know.

AI is not likely to take over the world, but will probably make major inroads in the enterprise IT space. This will happen as soon as business decision makers can be shown a compelling economic reason to adopt such technology. What is needed are use cases that lay out the expected ROIs for such investments. This paper examines the state of AI business rationalization, and looks at one company, IBM, that is attempting to provide support to business leaders presented with the decision to adopt AI solutions. Artificial intelligence (AI) has been very prevalent in the popular press, and seems to be comprised of equal parts speculation and hyperbole.

It's a little known but often dangerous problem: each year, 3,500 people in the U.S. -- mostly young children -- swallow button batteries. Normally, these batteries pass through the body without incident. But if they come into prolonged contact with esophagus or stomach tissue, the results can be harmful: the batteries can cause an electric current that produces hydroxide, which burns through body tissue. A postdoctoral student at the Massachusetts Institute of Technology, Shuhei Miyashita brought up this hazard to Daniela Rus, the professor who leads the Computer Science and Artificial Intelligence Laboratory. A simple experiment proved how hazardous the little batteries can be.

General Motors has completed its acquisition of Cruise Automation, the 3-year-old San Francisco startup that may provide a critical piece of technology in the quest to bring a fully autonomous car to market soon. The automaker won't disclose the final price or other terms until this summer when it reports second-quarter financial results. Multiple media sources have reported that GM is paying at least 1 billion for Cruise. "General Motors is pleased to announce it has completed the closing of the acquisition of Cruise Automation," the company said in a statement. The deal's consummation had been threatened by a pair of lawsuits involving Cruise cofounder and CEO Kyle Vogt and Jeremy Guillory, who said he was an early partner in the company.