Discount Factor as a Regularizer in Reinforcement Learning
This work addresses the problem of improving RL performance in limited data regimes for researchers and practitioners, though it is incremental as it formalizes a known but unstudied effect.
The paper investigates how using a lower discount factor in reinforcement learning acts as a regularizer, showing an explicit equivalence to adding a regularization term in Temporal-Difference methods and empirically comparing it to standard L2 regularization across various domains.
Specifying a Reinforcement Learning (RL) task involves choosing a suitable planning horizon, which is typically modeled by a discount factor. It is known that applying RL algorithms with a lower 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 several Temporal-Difference (TD) learning methods, we show an explicit equivalence between using a reduced discount factor and adding an explicit regularization term to the algorithm's loss. Motivated by the equivalence, we empirically study this technique compared to standard $L_2$ regularization by extensive experiments in discrete and continuous domains, using tabular and functional representations. Our experiments suggest the regularization effectiveness is strongly related to properties of the available data, such as size, distribution, and mixing rate.