Humans are not Boltzmann Distributions: Challenges and Opportunities for Modelling Human Feedback and Interaction in Reinforcement Learning
It addresses the problem of designing robust AI systems that interact with humans, but it is incremental as it critiques existing approaches without presenting new methods or results.
The paper argues that current reinforcement learning models oversimplify human feedback as rational and unbiased, calling for more realistic models that are personal, contextual, and dynamic to improve human-in-the-loop systems.
Reinforcement learning (RL) commonly assumes access to well-specified reward functions, which many practical applications do not provide. Instead, recently, more work has explored learning what to do from interacting with humans. So far, most of these approaches model humans as being (nosily) rational and, in particular, giving unbiased feedback. We argue that these models are too simplistic and that RL researchers need to develop more realistic human models to design and evaluate their algorithms. In particular, we argue that human models have to be personal, contextual, and dynamic. This paper calls for research from different disciplines to address key questions about how humans provide feedback to AIs and how we can build more robust human-in-the-loop RL systems.