Humans learn generalizable representations through efficient coding
Year: 2025
Type: article
Abstract: Reinforcement learning theory explains human behavior as driven by the goal of maximizing reward. Conventional approaches, however, offer limited insights into how people generalize from past experiences to new situations. Here, we propose refining the classical reinforcement learning framework by incorporating an efficient coding principle, which emphasizes maximizing reward using the simplest necessary representations. This refined framework predicts that intelligent agents, constrained by simpler representations, will inevitably: 1) distill environmental stimuli into fewer, abstract internal states, and 2) detect and utilize rewarding environmental features. Consequently, complex stimuli are mapped to compact representations, forming the foundation for generalization. We tested this idea in two experiments that examined human generalization. Our findings reveal that while conventional models fall short in generalization, models incorporating efficient coding achieve human-level performance. We argue that the classical RL objective, augmented with efficient coding, represents a more comprehensive computational framework for understanding human behavior in both learning and generalization. Reinforcement learning models of human behavior are limited in explaining the capacity for generalization. Here, the authors propose an efficient coding principle for reinforcement learning, whereby agents use compact representations, enabling human-like generalization.
Source: Nature Communications
Authors: Zeming Fang, Chris R. Sims
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $6,990
Award: G632850268
OpenAlex ID: W4409877367