Learning Task-Agnostic Skill Bases to Uncover Motor Primitives in Animal Behaviors

2025-06-18

AI Summary

This work addresses the oversimplification inherent in the discrete syllable assumption commonly adopted in animal behavior modeling. We propose a skill-basis-driven imitation learning framework that unsupervisedly disentangles reusable, task-agnostic continuous motor primitives from kinematic trajectories. Methodologically, we introduce the first integration of reinforcement learning with transition-probability-based representation learning to enable interpretable, continuously evolving skill-basis inference. Our approach advances beyond traditional discrete behavioral syllable paradigms through three core innovations: (i) skill-basis representation learning, (ii) dynamically mixed policy parameterization, and (iii) generative trajectory modeling. Extensive evaluation across grid-world environments, maze simulations, and real-world free-behavior video data demonstrates that our method extracts highly transferable skill components and generates more realistic, temporally coherent, and generalizable behavioral trajectories.

Technology Category

• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Intelligent Robots: Behavior Learning & Control
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Animals flexibly recombine a finite set of core motor primitives to meet diverse task demands, but existing behavior-segmentation methods oversimplify this process by imposing discrete syllables under restrictive generative assumptions. To reflect the animal behavior generation procedure, we introduce skill-based imitation learning (SKIL) for behavior understanding, a reinforcement learning-based imitation framework that (1) infers interpretable skill sets, i.e., latent basis functions of behavior, by leveraging representation learning on transition probabilities, and (2) parameterizes policies as dynamic mixtures of these skills. We validate our approach on a simple grid world, a discrete labyrinth, and unconstrained videos of freely moving animals. Across tasks, it identifies reusable skill components, learns continuously evolving compositional policies, and generates realistic trajectories beyond the capabilities of traditional discrete models. By exploiting generative behavior modeling with compositional representations, our method offers a concise, principled account of how complex animal behaviors emerge from dynamic combinations of fundamental motor primitives.

Problem

• Identify reusable motor primitives in animal behaviors
• Learn dynamic skill mixtures for behavior policies
• Generate realistic trajectories beyond discrete models

Innovation

• Uses reinforcement learning for imitation
• Infers interpretable skill sets
• Parameterizes policies as skill mixtures