An integrated approach is proposed across visual and textual data to both determine and justify a medical diagnosis by a neural network. As deep learning techniques improve, interest grows to apply them in medical applications. To enable a transition to workflows in a medical context that are aided by machine learning, the need exists for such algorithms to help justify the obtained outcome so human clinicians can judge their validity. In this work, deep learning methods are used to map a frontal X-Ray image to a continuous textual representation. This textual representation is decoded into a diagnosis and the associated textual justification that will help a clinician evaluate the outcome. Additionally, more explanatory data is provided for the diagnosis by generating a realistic X-Ray that belongs to the nearest alternative diagnosis. With a clinical expert opinion study on a subset of the X-Ray data set from the Indiana University hospital network, we demonstrate that our justification mechanism significantly outperforms existing methods that use saliency maps. While performing multi-task training with multiple loss functions, our method achieves excellent diagnosis accuracy and captioning quality when compared to current state-of-the-art single-task methods.

Time series are often associated with missing values, for instance due to faulty measurement devices, partially observed states, or costly measurement procedures [15]. These missing values impair the usefulness and interpretability of the data, leading to the problem of data imputation: estimating those missing values from the observed ones [38]. Multivariate time series, consisting of multiple correlated univariate time series or channels, give rise to two distinct ways of imputing missing information: (1) by exploiting temporal correlations within each channel, and (2) by exploiting correlations across channels, for example by using lowerdimensional representations of the data. For instance in a medical setting, if the blood pressure of a patient is unobserved, it can be informative that the heart rate at the current time is higher than normal and that the blood pressure was also elevated an hour ago. An ideal imputation model for multivariate time series should therefore take both of these sources of information into account.

The performance of imitation learning is typically upper-bounded by the performance of the demonstrator. Recent empirical results show that imitation learning via ranked demonstrations allows for better-than-demonstrator performance; however, ranked demonstrations may be difficult to obtain, and little is known theoretically about when such methods can be expected to outperform the demonstrator. To address these issues, we first contribute a sufficient condition for when better-than-demonstrator performance is possible and discuss why ranked demonstrations can contribute to better-than-demonstrator performance. Building on this theory, we then introduce Disturbance-based Reward Extrapolation (D-REX), a ranking-based imitation learning method that injects noise into a policy learned through behavioral cloning to automatically generate ranked demonstrations. By generating rankings automatically, ranking-based imitation learning can be applied in traditional imitation learning settings where only unlabeled demonstrations are available. We empirically validate our approach on standard MuJoCo and Atari benchmarks and show that D-REX can utilize automatic rankings to significantly surpass the performance of the demonstrator and outperform standard imitation learning approaches. D-REX is the first imitation learning approach to achieve significant extrapolation beyond the demonstrator's performance without additional side-information or supervision, such as rewards or human preferences.

We study a formalization of the grammar induction problem that models sentences as being generated by a compound probabilistic context-free grammar. In contrast to traditional formulations which learn a single stochastic grammar, our context-free rule probabilities are modulated by a per-sentence continuous latent variable, which induces marginal dependencies beyond the traditional context-free assumptions. Inference in this grammar is performed by collapsed variational inference, in which an amortized variational posterior is placed on the continuous variable, and the latent trees are marginalized with dynamic programming. Experiments on English and Chinese show the effectiveness of our approach compared to recent state-of-the-art methods for grammar induction from words with neural language models.

We propose symbolic learning as extensions to standard inductive learning models such as neural nets as a means to solve few shot learning problems. We device a class of visual discrimination puzzles that calls for recognizing objects and object relationships as well learning higher-level concepts from very few images. We propose a two-phase learning framework that combines models learned from large data sets using neural nets and symbolic first-order logic formulas learned from a few shot learning instance. We develop first-order logic synthesis techniques for discriminating images by using symbolic search and logic constraint solvers. By augmenting neural nets with them, we develop and evaluate a tool that can solve few shot visual discrimination puzzles with interpretable concepts.

Imitation learning, followed by reinforcement learning algorithms, is a promising paradigm to solve complex control tasks sample-efficiently. However, learning from demonstrations often suffers from the covariate shift problem, which results in cascading errors of the learned policy. We introduce a notion of conservatively-extrapolated value functions, which provably lead to policies with self-correction. We design an algorithm Value Iteration with Negative Sampling (VINS) that practically learns such value functions with conservative extrapolation. We show that VINS can correct mistakes of the behavioral cloning policy on simulated robotics benchmark tasks. We also propose the algorithm of using VINS to initialize a reinforcement learning algorithm, which is shown to outperform significantly prior works in sample efficiency.

This paper addresses two challenges facing sampling-based kinodynamic motion planning: a way to identify good candidate states for local transitions and the subsequent computationally intractable steering between these candidate states. Through the combination of sampling-based planning, a Rapidly Exploring Randomized Tree (RRT) and an efficient kinodynamic motion planner through machine learning, we propose an efficient solution to long-range planning for kinodynamic motion planning. First, we use deep reinforcement learning to learn an obstacle-avoiding policy that maps a robot's sensor observations to actions, which is used as a local planner during planning and as a controller during execution. Second, we train a reachability estimator in a supervised manner, which predicts the RL policy's time to reach a state in the presence of obstacles. Lastly, we introduce RL-RRT that uses the RL policy as a local planner, and the reachability estimator as the distance function to bias tree-growth towards promising regions. We evaluate our method on three kinodynamic systems, including physical robot experiments. Results across all three robots tested indicate that RL-RRT outperforms state of the art kinodynamic planners in efficiency, and also provides a shorter path finish time than a steering function free method. The learned local planner policy and accompanying reachability estimator demonstrate transferability to the previously unseen experimental environments, making RL-RRT fast because the expensive computations are replaced with simple neural network inference. Video: https://youtu.be/dDMVMTOI8KY

During one experiment, the poker bot Pluribus played against five professional players. During one experiment, the poker bot Pluribus played against five professional players. In artificial intelligence, it's a milestone when a computer program can beat top players at a game like chess. But a game like poker, specifically six-player Texas Hold'em, has been too tough for a machine to master -- until now. Researchers say they have designed a bot called Pluribus capable of taking on poker professionals in the most popular form of poker and winning.

Robots helped build your car and pack your latest online shopping order. A chatbot might help you figure out your credit card balance. A computer program might scan and process your resume when you apply for work. What will work in America look like a decade for now? A team of economists at the McKinsey Global Institute set off to figure out in a new report out Thursday.

A computer has beaten five of the world's champion players at poker -- a game once thought too difficult for machines to master. It is the latest milestone marking the superior powers of machines over people and the first time a computer has beaten more than one opponent in a complex game of strategy and calculation. Computers first defeated the human world champion at chess in 1996 -- and the even-more complex Chinese strategy game of Go two years ago. But poker has posed a tougher challenge as it involves several players around the table. And unlike in chess or Go, the computer does not have access to all the information available as it cannot see its opponent's cards.