Jul 12, 2020
Specifying a Reinforcement Learning (RL) task involves choosing a suitable planning horizon, which is typically modeled by an evaluation discount factor. It is known that applying RL algorithms with a discount set lower than the evaluation discount factor can act as a regularizer, improving performance in the limited data regime. Yet the exact nature of this regularizer has not been investigated. In this work, we fill in this gap. For TD learning and expected SARSA, we show an explicit equivalence between using a reduced discount factor and adding an explicit regularization term to the algorithm loss. For a fixed policy, we argue that chains with a uniform stationary distribution and a fast mixing rate are amenable to regularization with a reduced discount. We validate this conclusion with extensive experiments in discrete and continuous domains, using tabular and functional representations.
The International Conference on Machine Learning (ICML) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence known as machine learning. ICML is globally renowned for presenting and publishing cutting-edge research on all aspects of machine learning used in closely related areas like artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, and robotics. ICML is one of the fastest growing artificial intelligence conferences in the world. Participants at ICML span a wide range of backgrounds, from academic and industrial researchers, to entrepreneurs and engineers, to graduate students and postdocs.
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Presentations on similar topic, category or speaker