Reannealing of Decaying Exploration Based On Heuristic Measure in Deep Q-Network
arXiv:2009.14297
Abstract
Existing exploration strategies in reinforcement learning (RL) often either ignore the history or feedback of search, or are complicated to implement. There is also a very limited literature showing their effectiveness over diverse domains. We propose an algorithm based on the idea of reannealing, that aims at encouraging exploration only when it is needed, for example, when the algorithm detects that the agent is stuck in a local optimum. The approach is simple to implement. We perform an illustrative case study showing that it has potential to both accelerate training and obtain a better policy.
References in corpus (6)
- Adam: A Method for Stochastic Optimization
- #Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning
- Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models
- Efficient approaches for escaping higher order saddle points in non-convex optimization
- UCB Exploration via Q-Ensembles
- A Survey of Online Experiment Design with the Stochastic Multi-Armed Bandit