Randomized Ensembled Double Q-Learning: Learning Fast Without a Model
arXiv:2101.05982
Abstract
Using a high Update-To-Data (UTD) ratio, model-based methods have recently achieved much higher sample efficiency than previous model-free methods for continuous-action DRL benchmarks. In this paper, we introduce a simple model-free algorithm, Randomized Ensembled Double Q-Learning (REDQ), and show that its performance is just as good as, if not better than, a state-of-the-art model-based algorithm for the MuJoCo benchmark. Moreover, REDQ can achieve this performance using fewer parameters than the model-based method, and with less wall-clock run time. REDQ has three carefully integrated ingredients which allow it to achieve its high performance: (i) a UTD ratio >> 1; (ii) an ensemble of Q functions; (iii) in-target minimization across a random subset of Q functions from the ensemble. Through carefully designed experiments, we provide a detailed analysis of REDQ and related model-free algorithms. To our knowledge, REDQ is the first successful model-free DRL algorithm for continuous-action spaces using a UTD ratio >> 1.
Published as a conference paper at ICLR 2021
References in corpus (7)
- Benchmarking Model-Based Reinforcement Learning
- Reproducibility of Benchmarked Deep Reinforcement Learning Tasks for Continuous Control
- Averaged-DQN: Variance Reduction and Stabilization for Deep Reinforcement Learning
- Model-Augmented Actor-Critic: Backpropagating through Paths
- Bidirectional Model-based Policy Optimization
- The Effect of Multi-step Methods on Overestimation in Deep Reinforcement Learning
- Maximum Entropy Model Rollouts: Fast Model Based Policy Optimization without Compounding Errors