Fast deep reinforcement learning using online adjustments from the past
arXiv:1810.08163
Abstract
We propose Ephemeral Value Adjusments (EVA): a means of allowing deep reinforcement learning agents to rapidly adapt to experience in their replay buffer. EVA shifts the value predicted by a neural network with an estimate of the value function found by planning over experience tuples from the replay buffer near the current state. EVA combines a number of recent ideas around combining episodic memory-like structures into reinforcement learning agents: slot-based storage, content-based retrieval, and memory-based planning. We show that EVAis performant on a demonstration task and Atari games.
Accepted at NIPS 2018
Cited by in corpus (7)
- OpenSpiel: A Framework for Reinforcement Learning in Games
- Sample-Efficient Deep Reinforcement Learning via Episodic Backward Update
- Learning to Learn Variational Semantic Memory
- Self-Imitation Advantage Learning
- Model-Based Episodic Memory Induces Dynamic Hybrid Controls
- Random Projection in Neural Episodic Control
- Episodic Memory for Learning Subjective-Timescale Models