2019/11/21 by Marta Sarrico, Sarrico, Marta, Kai Arulkumaran +9
Computer Science · Decision Sciences · Mathematics · #Advanced Bandit Algorithms Research #Adversarial Robustness in Machine Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Neural and Evolutionary Computing (cs.NE) #Reinforcement Learning in Robotics #cs.LG #cs.NE #stat.ML
paper · pdf · doi:10.48550/arxiv.1911.09615
Workshop on Biological and Artificial Reinforcement Learning, NeurIPS 2019
arxiv created 2019/11/21 · openalex publication_date 2019/11/21 · arxiv updated 2019/11/22 · openalex created_date 2019/12/05 · openalex updated_date 2026/07/28
Deep networks have enabled reinforcement learning to scale to more complex and challenging domains, but these methods typically require large quantities of training data. An alternative is to use sample-efficient episodic control methods: neuro-inspired algorithms which use non-/semi-parametric models that predict values based on storing and retrieving previously experienced transitions. One way to further improve the sample efficiency of these approaches is to use more principled exploration strategies. In this work, we therefore propose maximum entropy mellowmax episodic control (MEMEC), which samples actions according to a Boltzmann policy with a state-dependent temperature. We demonstrate that MEMEC outperforms other uncertainty- and softmax-based exploration methods on classic reinforcement learning environments and Atari games, achieving both more rapid learning and higher final rewards.