Best-Response Bayesian Reinforcement Learning with Bayes-adaptive POMDPs for Centaurs

Çelikok, Mustafa Mert, Oliehoek, Frans A ORCID: 0000-0003-4372-5055 and Kaski, Samuel
(2022) Best-Response Bayesian Reinforcement Learning with Bayes-adaptive POMDPs for Centaurs. [Preprint]

[thumbnail of 2204.01160v1.pdf] Text
2204.01160v1.pdf - Submitted version

Download (1MB) | Preview


Centaurs are half-human, half-AI decision-makers where the AI's goal is to complement the human. To do so, the AI must be able to recognize the goals and constraints of the human and have the means to help them. We present a novel formulation of the interaction between the human and the AI as a sequential game where the agents are modelled using Bayesian best-response models. We show that in this case the AI's problem of helping bounded-rational humans make better decisions reduces to a Bayes-adaptive POMDP. In our simulated experiments, we consider an instantiation of our framework for humans who are subjectively optimistic about the AI's future behaviour. Our results show that when equipped with a model of the human, the AI can infer the human's bounds and nudge them towards better decisions. We discuss ways in which the machine can learn to improve upon its own limitations as well with the help of the human. We identify a novel trade-off for centaurs in partially observable tasks: for the AI's actions to be acceptable to the human, the machine must make sure their beliefs are sufficiently aligned, but aligning beliefs might be costly. We present a preliminary theoretical analysis of this trade-off and its dependence on task structure.

Item Type: Preprint
Additional Information: This paper is presented in part at the International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS) 2022
Uncontrolled Keywords: cs.AI, cs.AI, cs.LG, cs.MA
Divisions: Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science
Depositing User: Symplectic Admin
Date Deposited: 19 Apr 2022 15:09
Last Modified: 24 Mar 2024 19:08
DOI: 10.48550/arxiv.2204.01160
Related URLs: