Evolving Programmatic Skill Networks

TL;DR

This paper introduces the Programmatic Skill Network (PSN), a framework for continual skill acquisition in embodied environments, leveraging large language models for structured skill learning, reuse, and adaptation.

Contribution

The paper presents PSN, a novel symbolic, compositional framework for evolving skill networks with mechanisms for fault localization, stability, and structural refactoring, inspired by neural training dynamics.

Findings

PSN enables robust skill reuse and rapid adaptation.

Demonstrates strong generalization across diverse tasks.

Maintains network compactness through structural refactoring.

Abstract

We study continual skill acquisition in open-ended embodied environments where an agent must construct, refine, and reuse an expanding library of executable skills. We introduce the Programmatic Skill Network (PSN), a framework in which skills are executable symbolic programs forming a compositional network that evolves through experience. PSN defines three core mechanisms instantiated via large language models: (1)REFLECT for structured fault localization over skill compositions, (2) progressive optimization with maturity-aware update gating that stabilizes reliable skills while maintaining plasticity for uncertain ones, and (3) canonical structural refactoring under rollback validation that maintains network compactness. We further show that PSN's learning dynamics exhibit structural parallels to neural network training. Experiments on MineDojo and Crafter demonstrate robust skill reuse, rapid adaptation, and strong generalization across open-ended task distributions.\footnote{We plan to open-source the code.