Complex behavior from intrinsic motivation to occupy action-state path space
This work offers a foundational theory for understanding intrinsic motivation in AI and animal behavior, moving beyond traditional reward-based models.
The authors tackled the problem of explaining complex, reward-free behaviors in agents by proposing a maximum occupancy principle, where the goal is to maximize future action-state path occupancy rather than reward maximization, and demonstrated that this principle generates behaviors like dancing and altruism in both discrete and continuous tasks.
Most theories of behavior posit that agents tend to maximize some form of reward or utility. However, animals very often move with curiosity and seem to be motivated in a reward-free manner. Here we abandon the idea of reward maximization, and propose that the goal of behavior is maximizing occupancy of future paths of actions and states. According to this maximum occupancy principle, rewards are the means to occupy path space, not the goal per se; goal-directedness simply emerges as rational ways of searching for resources so that movement, understood amply, never ends. We find that action-state path entropy is the only measure consistent with additivity and other intuitive properties of expected future action-state path occupancy. We provide analytical expressions that relate the optimal policy and state-value function, and prove convergence of our value iteration algorithm. Using discrete and continuous state tasks, including a high--dimensional controller, we show that complex behaviors such as `dancing', hide-and-seek and a basic form of altruistic behavior naturally result from the intrinsic motivation to occupy path space. All in all, we present a theory of behavior that generates both variability and goal-directedness in the absence of reward maximization.