Paper Info
Title
RL$^2$: Fast Reinforcement Learning via Slow Reinforcement Learning.
Abstract
Deep reinforcement learning (deep RL) has been successful in learning sophisticated behaviors automatically; however, the learning process requires a huge number of trials. In contrast, animals can learn new tasks in just a few trials, benefiting from their prior knowledge about the world. This paper seeks to bridge this gap. Rather than designing a “fast” reinforcement learning algorithm, we propose to represent it as a recurrent neural network (RNN) and learn it from data. In our proposed method, RL^2, the algorithm is encoded in the weights of the RNN, which are learned slowly through a general-purpose (“slow”) RL algorithm. The RNN receives all information a typical RL algorithm would receive, including observations, actions, rewards, and termination flags; and it retains its state across episodes in a given Markov Decision Process (MDP). The activations of the RNN store the state of the “fast” RL algorithm on the current (previously unseen) MDP. We evaluate RL^2 experimentally on both small-scale and large-scale problems. On the small-scale side, we train it to solve randomly generated multi-arm bandit problems and finite MDPs. After RL^2 is trained, its performance on new MDPs is close to human-designed algorithms with optimality guarantees. On the large-scale side, we test RL^2 on a vision-based navigation task and show that it scales up to high-dimensional problems.
Year
Venue
Field
2016
arXiv: Artificial Intelligence
Computer science,Markov decision process,Q-learning,Recurrent neural network,Artificial intelligence,Deep learning,Reinforcement learning algorithm,Error-driven learning,Machine learning,Learning classifier system,Reinforcement learning
DocType
Volume
Citations 
Journal
abs/1611.02779
0
PageRank 
References 
Authors
0.34
0
6
Name
Order
Citations
PageRank
Yan Duan177527.97
John Schulman2180666.95
Xi Chen3164954.94
Peter L. Bartlett454821039.97
Ilya Sutskever5258141120.24
Pieter Abbeel66363376.48