S
Scott Fujimoto
Researcher at McGill University
Publications - 19
Citations - 4164
Scott Fujimoto is an academic researcher from McGill University. The author has contributed to research in topics: Reinforcement learning & Computer science. The author has an hindex of 9, co-authored 16 publications receiving 2232 citations.
Papers
More filters
Posted Content
Addressing Function Approximation Error in Actor-Critic Methods
TL;DR: This paper builds on Double Q-learning, by taking the minimum value between a pair of critics to limit overestimation, and draws the connection between target networks and overestimation bias.
Proceedings Article
Addressing Function Approximation Error in Actor-Critic Methods
TL;DR: In this paper, the authors show that the overestimation bias persists in an actor-critic setting and propose novel mechanisms to minimize its effects on both the actor and the critic.
Proceedings Article
Off-Policy Deep Reinforcement Learning without Exploration
TL;DR: This paper introduces a novel class of off-policy algorithms, batch-constrained reinforcement learning, which restricts the action space in order to force the agent towards behaving close to on-policy with respect to a subset of the given data.
Posted Content
Benchmarking Batch Deep Reinforcement Learning Algorithms.
TL;DR: This paper benchmark the performance of recent off-policy and batch reinforcement learning algorithms under unified settings on the Atari domain, with data generated by a single partially-trained behavioral policy, and finds that many of these algorithms underperform DQN trained online with the same amount of data.
Proceedings Article
GEOMetrics: Exploiting Geometric Structure for Graph-Encoded Objects
TL;DR: In this paper, a graph convolutional update preserving vertex information and an adaptive splitting heuristic allowing detail to emerge is proposed for 3D object reconstruction from images with the ShapeNet dataset.