Cooperative Inverse Reinforcement Learning
This addresses the challenge of ensuring AI systems align with human values, which is crucial for safety and helpfulness in human-robot interactions, representing a novel foundational approach rather than an incremental improvement.
The paper tackles the value alignment problem for autonomous systems by proposing cooperative inverse reinforcement learning (CIRL), a game-theoretic framework where a robot learns a human's reward function through cooperative interaction, leading to behaviors like active teaching that improve alignment compared to classical methods.
For an autonomous system to be helpful to humans and to pose no unwarranted risks, it needs to align its values with those of the humans in its environment in such a way that its actions contribute to the maximization of value for the humans. We propose a formal definition of the value alignment problem as cooperative inverse reinforcement learning (CIRL). A CIRL problem is a cooperative, partial-information game with two agents, human and robot; both are rewarded according to the human's reward function, but the robot does not initially know what this is. In contrast to classical IRL, where the human is assumed to act optimally in isolation, optimal CIRL solutions produce behaviors such as active teaching, active learning, and communicative actions that are more effective in achieving value alignment. We show that computing optimal joint policies in CIRL games can be reduced to solving a POMDP, prove that optimality in isolation is suboptimal in CIRL, and derive an approximate CIRL algorithm.