This paper proposes REFUEL, a reinforcement learning method with two techniques: reward shaping and feature rebuilding, to improve the performance of online symptom checking for disease diagnosis.

The growing importance of massive datasets used for deep learning makes robustness to label noise a critical property for classifiers to have.

Estimating individual treatment effect (ITE) is a challenging problem in causal inference, due to the missing counterfactuals and the selection bias. Existing ITE estimation methods mainly focus on balancing the distributions of control and treated groups, but ignore the local similarity information that provides meaningful constraints on the ITE estimation. In this paper, we propose a local similarity preserved i ndividual t reatment effect (SITE) estimation method based on deep representation learning. SITE preserves local similarity and balances data distributions simultaneously, by focusing on several hard samples in each mini-batch. Experimental results on synthetic and three real-world datasets demonstrate the advantages of the proposed SITE method, compared with the state-of-the-art ITE estimation methods.

Neural Information Processing Systems http://nips.cc/

The classic Mallows model is a widely-used tool to realize distributions on permutations.

Though the context is chosen arbitrarily or adversarially, the reward is assumed to be a stochastic function of a feature vector that encodes the context and selected action. Our goal is to devise private learners for the contextual linear bandit problem.

This paper studies the problem of distributed stochastic optimization in an adversarial setting where, out of m machines which allegedly compute stochastic gradients every iteration, an -fraction are Byzantine, and may behave adversari-ally. Our main result is a variant of stochastic gradient descent (SGD) which finds " -approximate minimizers of convex functions in T = e O

Neural Information Processing Systems http://nips.cc/

Batch Normalization (BatchNorm) is a widely adopted technique that enables faster and more stable training of deep neural networks (DNNs). Despite its pervasiveness, the exact reasons for BatchNorm's effectiveness are still poorly understood. The popular belief is that this effectiveness stems from controlling the change of the layers' input distributions during training to reduce the so-called internal covariate shift. In this work, we demonstrate that such distributional stability of layer inputs has little to do with the success of BatchNorm. Instead, we uncover a more fundamental impact of BatchNorm on the training process: it makes the optimization landscape significantly smoother. This smoothness induces a more predictive and stable behavior of the gradients, allowing for faster training.

There are two main classes of approaches to explain the prediction of a model.