RL Unplugged: A Collection of Benchmarks for Offline Reinforcement Learning.
Çaglar GülçehreZiyu WangAlexander NovikovThomas PaineSergio Gómez ColmenarejoKonrad ZolnaRishabh AgarwalJosh MerelDaniel J. MankowitzCosmin PaduraruGabriel Dulac-ArnoldJerry LiMohammad NorouziMatthew HoffmanNicolas HeessNando de FreitasPublished in: NeurIPS (2020)
Keyphrases
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- model free
- state space
- temporal difference
- control problems
- real time
- optimal policy
- partially observable
- learning process
- rl algorithms
- machine learning
- direct policy search
- function approximators
- action space
- action selection
- temporal difference learning
- partially observable domains
- markov decision processes
- supervised learning
- multi agent
- learning algorithm
- neural network
- optimal control
- markov decision process
- transfer learning
- continuous state
- actor critic
- td learning