165 citations · 439 across the 10 of their papers we have counts for
3 papers · 1 filter
Models, Pixels, and Rewards: Evaluating Design Trade-offs in Visual Model-Based Reinforcement Learning
Mohammad Babaeizadeh, Mohammad Taghi Saffar, Danijar Hafner +4
Model-based reinforcement learning (MBRL) methods have shown strong sample efficiency and performance across a variety of tasks, including when faced with high-dimensional visual o…
A Benchmark for Interpretability Methods in Deep Neural Networks
Sara Hooker, Dumitru Erhan, Pieter-Jan Kindermans +1
We propose an empirical measure of the approximate accuracy of feature importance estimates in deep neural networks. Our results across several large-scale image classification dat…
Sample-efficient Nonstationary Policy Evaluation for Contextual Bandits
Miroslav Dudik, Dumitru Erhan, John Langford +1
We present and prove properties of a new offline policy evaluator for an exploration learning setting which is superior to previous evaluators. In particular, it simultaneously and…