Inverse reinforcement learning (IRL) attempts to infer human rewards or preferences from observed behavior. Since human planning systematically deviates from rationality, several approaches have been tried to account for specific human shortcomings. However, the general problem of inferring the reward function of an agent of unknown rationality has received little attention. Unlike the well-known ambiguity problems in IRL, this one is practically relevant but cannot be resolved by observing the agent's policy in enough environments. This paper shows (1) that a No Free Lunch result implies it is impossible to uniquely decompose a policy into a planning algorithm and reward function, and (2) that even with a reasonable simplicity prior/Occam's razor on the set of decompositions, we cannot distinguish between the true decomposition and others that lead to high regret. To address this, we need simple `normative' assumptions, which cannot be deduced exclusively from observations.
Stuart Armstrong (Oxford University)
Sören Mindermann (Vector Institute)
More from the Same Authors
2020 Poster: Identifying Causal-Effect Inference Failure with Uncertainty-Aware Models »
Andrew Jesson · Sören Mindermann · Uri Shalit · Yarin Gal
2020 : How Robust are the Estimated Effects of Nonpharmaceutical Interventions against COVID-19? »
Sören Mindermann · Mrinank Sharma · Jan Brauner