Workshop: Responsible Decision Making in Dynamic Environments

Reward Reports for Reinforcement Learning

Thomas Krendl Gilbert · Sarah Dean · Nathan Lambert · Tom Zick · Aaron Snoswell


The desire to build good systems in the face of complex societal effects requires a dynamic approach towards equity and access. Recent approaches to machine learning (ML) documentation have demonstrated the promise of discursive frameworks for deliberation about these complexities. However, these developments have been grounded in a static ML paradigm, leaving the role of feedback and post-deployment performance unexamined. Meanwhile, recent work in reinforcement learning design has shown that the effects of optimization objectives on the resultant system behavior can be wide-ranging and unpredictable. In this paper we sketch a framework for documenting deployed learning systems, which we call \textit{Reward Reports}.

Chat is not available.