Towards a Research Community in Interpretable Reinforcement Learning: the InterpPol Workshop
This work is incremental as it focuses on community-building rather than advancing technical solutions in interpretable RL.
The paper tackles the need for a dedicated research community in interpretable reinforcement learning by proposing the InterpPol Workshop, which aims to address key questions about explainability, interpretability, and methods in RL without providing specific results or numbers.
Embracing the pursuit of intrinsically explainable reinforcement learning raises crucial questions: what distinguishes explainability from interpretability? Should explainable and interpretable agents be developed outside of domains where transparency is imperative? What advantages do interpretable policies offer over neural networks? How can we rigorously define and measure interpretability in policies, without user studies? What reinforcement learning paradigms,are the most suited to develop interpretable agents? Can Markov Decision Processes integrate interpretable state representations? In addition to motivate an Interpretable RL community centered around the aforementioned questions, we propose the first venue dedicated to Interpretable RL: the InterpPol Workshop.