Temporally-Extended ε-Greedy Exploration
arXiv:2006.01782
Abstract
Recent work on exploration in reinforcement learning (RL) has led to a series of increasingly complex solutions to the problem. This increase in complexity often comes at the expense of generality. Recent empirical studies suggest that, when applied to a broader set of domains, some sophisticated exploration methods are outperformed by simpler counterparts, such as ε-greedy. In this paper we propose an exploration algorithm that retains the simplicity of ε-greedy while reducing dithering. We build on a simple hypothesis: the main limitation of ε-greedy exploration is its lack of temporal persistence, which limits its ability to escape local optima. We propose a temporally extended form of ε-greedy that simply repeats the sampled action for a random duration. It turns out that, for many duration distributions, this suffices to improve exploration on a large set of domains. Interestingly, a class of distributions inspired by ecological models of animal foraging behaviour yields particularly strong performance.
References in corpus (18)
- Dueling Network Architectures for Deep Reinforcement Learning
- Noisy Networks for Exploration
- DeepMind Control Suite
- Rainbow: Combining Improvements in Deep Reinforcement Learning
- Exploration by Random Network Distillation
- Agent57: Outperforming the Atari Human Benchmark
- (More) Efficient Reinforcement Learning via Posterior Sampling
- Go-Explore: a New Approach for Hard-Exploration Problems
- A Bayesian Sampling Approach to Exploration in Reinforcement Learning
- Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning
- Never Give Up: Learning Directed Exploration Strategies
- A Laplacian Framework for Option Discovery in Reinforcement Learning
- Learning to Repeat: Fine Grained Action Repetition for Deep Reinforcement Learning
- Successor Uncertainties: Exploration and Uncertainty in Temporal Difference Learning
- On Bonus-Based Exploration Methods in the Arcade Learning Environment
- The Termination Critic
- Discovering Options for Exploration by Minimizing Cover Time
- When Simple Exploration is Sample Efficient: Identifying Sufficient Conditions for Random Exploration to Yield PAC RL Algorithms