arXiv (Cornell University) · 2018 · 70 citations · 0 references
ComputingAutonomous SystemsComputer ScienceArtificial IntelligenceMachine LearningRobot LearningAutonomous ExplorationExploration-exploitation TradeoffImitation LearningPolicy LearningReinforcement LearningReinforcement Learning FoundationsCount-based ExplorationExplorationExploration In Reinforcement LearningOff-policy LearningPolicy OptimizationProximal AlgorithmsProximal Policy Optimization
This paper proposes Self-Imitation Learning (SIL), a simple off-policy actor-critic algorithm that learns to reproduce the agent's past good decisions. This algorithm is designed to verify our hypothesis that exploiting past good experiences can indirectly drive deep exploration. Our empirical results show that SIL significantly improves advantage actor-critic (A2C) on several hard exploration Atari games and is competitive to the state-of-the-art count-based exploration methods. We also show that SIL improves proximal policy optimization (PPO) on MuJoCo tasks.