Upside-Down Reinforcement Learning for More Interpretable Optimal Control
This work addresses the need for more transparent and safe reinforcement learning, though it is incremental as it applies an existing method to new function approximators.
The paper investigates whether tree-based methods like Random Forests can replace Neural Networks in Upside-Down Reinforcement Learning, finding they perform comparably while offering more interpretable policies for optimal control benchmarks.
Model-Free Reinforcement Learning (RL) algorithms either learn how to map states to expected rewards or search for policies that can maximize a certain performance function. Model-Based algorithms instead, aim to learn an approximation of the underlying model of the RL environment and then use it in combination with planning algorithms. Upside-Down Reinforcement Learning (UDRL) is a novel learning paradigm that aims to learn how to predict actions from states and desired commands. This task is formulated as a Supervised Learning problem and has successfully been tackled by Neural Networks (NNs). In this paper, we investigate whether function approximation algorithms other than NNs can also be used within a UDRL framework. Our experiments, performed over several popular optimal control benchmarks, show that tree-based methods like Random Forests and Extremely Randomized Trees can perform just as well as NNs with the significant benefit of resulting in policies that are inherently more interpretable than NNs, therefore paving the way for more transparent, safe, and robust RL.