Linking Homeostasis to Reinforcement Learning: Internal State Control of Motivated Behavior

TL;DR

This paper introduces Homeostatically Regulated Reinforcement Learning (HRRL), a framework linking biological internal state regulation with computational learning, enabling adaptive, motivated behaviors in AI and biological systems.

Contribution

The paper proposes HRRL as a novel framework integrating biological principles with reinforcement learning to model internal state regulation and motivated behavior.

Findings

HRRL produces risk aversion and anticipatory regulation behaviors.

Extension to deep RL enables autonomous exploration and hierarchical behaviors.

HRRL offers a biologically plausible foundation for understanding motivation and decision-making.

Abstract

For living beings, survival depends on effective regulation of internal physiological states through motivated behaviors. In this perspective we propose that Homeostatically Regulated Reinforcement Learning (HRRL) as a framework to describe biological agents that optimize internal states via learned predictive control strategies, integrating biological principles with computational learning. We show that HRRL inherently produces multiple behaviors such as risk aversion, anticipatory regulation, and adaptive movement, aligning with observed biological phenomena. Its extension to deep reinforcement learning enables autonomous exploration, hierarchical behavior, and potential real-world robotic applications. We argue further that HRRL offers a biologically plausible foundation for understanding motivation, learning, and decision-making, with broad implications for artificial intelligence, neuroscience, and understanding the causes of psychiatric disorders, ultimately advancing our understanding of adaptive behavior in complex environments.