Offline Multi-agent Continual Cooperation via Skill Partition and Reuse

• URL: http://arxiv.org/abs/2606.25389v1
• Date: Wed, 24 Jun 2026 04:40:21 GMT
• Title: Offline Multi-agent Continual Cooperation via Skill Partition and Reuse
• Abstract summary: COMAD is a principled framework for Continual Offline Multi-agent Skill Discovery via Skill Partition and Reuse.<n>We first discover skills from mixed multi-agent behavior data with an auto-encoder to transform coordination knowledge into reusable coordination skills.<n>We construct a skill-augmented policy learning objective with multi-head architectures, explicitly guiding the advantage function with reusable skills identified.
• Score: 5.758813772286136
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Extracting skills from multi-agent offline dataset improves learning efficiency via sharing task-invariant coordination skills among tasks. In settings where tasks occur sequentially and the space of skills grows exponentially, existing approaches that rely on heuristically designed and fixed-sized skill libraries struggle to resolve the problem of distributional shift and interference, facing catastrophic forgetting and plasticity loss. To address this problem and endow agents with the ability to continually discover and reuse coordination skills in open-environment, we propose COMAD, a principled framework for Continual Offline Multi-agent Skill Discovery via Skill Partition and Reuse. We first discover skills from mixed multi-agent behavior data with an auto-encoder to transform coordination knowledge into reusable coordination skills. Then we construct a skill-augmented policy learning objective with multi-head architectures, explicitly guiding the advantage function with reusable skills identified via a density-based reusability estimator. Theoretical analysis shows our method approximates the optimum of a continual skill discovery problem. Empirical results across diverse MARL benchmarks show that COMAD continually expands its skill library to mitigate interference, achieving superior forward and backward transfer for task streams compared to multiple baselines.

Related papers

• SkillCommit: Evolving Agent Skills through Behaviorally Validated Scope Expansion [15.498568310060229]
  We propose SkillCommit, an online skill evolution framework that transforms experience into a hierarchical library of reusable skills.<n>As related skills accumulate, SkillCommit abstracts those sharing a common behavioral mechanism into higher-level skills.<n> Experiments on RuleArena, OpenExempt and KOR-Bench demonstrate that SkillCommit consistently improves agent performance across diverse domains.
  arXiv  Detail & Related papers  (2026-08-15T11:03:07Z)
• Rethinking Self-Evolution: A Constrained Exploration-Exploitation Process for Mitigating Skill Overfitting [20.811638832981178]
  Large language model (LLM) agents need to accumulate and reuse experience from past interactions.<n>Data-driven skill optimization is prone to overfitting to the limited trajectories collected from real environments.<n>We propose SkillBoost, a three-stage framework that mitigates both risks.
  arXiv  Detail & Related papers  (2026-07-29T09:05:40Z)
• Skill Self-Play: Pushing the Frontier of LLM Capability with Co-Evolving Skills [53.15216425100295]
  We introduce Skill Self-Play (Skill-SP), a co-evolutionary framework comprising a proposer, a solver, and a skill controller.<n>The proposer generates challenging tasks conditioned on dynamically sampled skills; the solver explores candidate solutions to push its capability boundaries; and the skill controller collects execution feedback to update and expand the skill library.<n>This interactive co-evolution effectively bridges the gap between structured verification and open-ended exploration.
  arXiv  Detail & Related papers  (2026-07-24T17:59:22Z)
• SkillComposer: Learning to Evolve Agent Skills for Specification and Generalization [99.18305770138677]
  Agent skills are reusable strategies that guide agent reasoning and action.<n>Current skill construction methods treat the problem as one-shot extraction.<n>We introduce SkillComposer, a framework that decomposes skill construction into three learnable operations.
  arXiv  Detail & Related papers  (2026-06-04T12:16:08Z)
• Swarm Skills: A Portable, Self-Evolving Multi-Agent System Specification for Coordination Engineering [50.72456732323037]
  We propose textbfSwarm Skills, a portable specification that extends the Anthropic Skills standard with multi-agent semantics.<n>Swarm Skills turns multi-agent into first-class, distributable assets that consist of roles, execution bounds, and a built-in semantic structure for self-evolution.<n>We show how Swarm Skills achieves zero-adapter cross-agent portability via progressive disclosure, enabling agent teams to self-evolve their coordination strategies without framework lock-in.
  arXiv  Detail & Related papers  (2026-05-11T06:26:58Z)
• SkillOS: Learning Skill Curation for Self-Evolving Agents [67.94374107466957]
  We propose SkillOS, an experience-driven RL training recipe for learning skill curation in self-evolving agents.<n>SkillOS pairs a frozen agent executor that retrieves and applies skills with a trainable skill curator that updates an external SkillRepo from accumulated experience.<n>We show that SkillOS consistently outperforms memory-free and strong memory-based baselines in both effectiveness and efficiency.
  arXiv  Detail & Related papers  (2026-05-07T17:31:50Z)
• Goal-Oriented Skill Abstraction for Offline Multi-Task Reinforcement Learning [25.18006424626525]
  GO-Skill is a novel approach designed to extract and utilize reusable skills to enhance knowledge transfer and task performance.<n>Our approach uncovers reusable skills through a goal-oriented skill extraction process and leverages vector quantization to construct a discrete skill library.<n>We integrate these skills using hierarchical policy learning, enabling the construction of a high-level policy that dynamically orchestrates discrete skills to accomplish specific tasks.
  arXiv  Detail & Related papers  (2025-07-09T07:54:49Z)
• Disentangled Unsupervised Skill Discovery for Efficient Hierarchical Reinforcement Learning [39.991887534269445]
  Disentangled Unsupervised Skill Discovery (DUSDi) is a method for learning disentangled skills that can be efficiently reused to solve downstream tasks. DUSDi decomposes skills into disentangled components, where each skill component only affects one factor of the state space. DUSDi successfully learns disentangled skills, and significantly outperforms previous skill discovery methods when it comes to applying the learned skills to solve downstream tasks.
  arXiv  Detail & Related papers  (2024-10-15T04:13:20Z)
• Skills Regularized Task Decomposition for Multi-task Offline Reinforcement Learning [11.790581500542439]
  Reinforcement learning (RL) with diverse offline datasets can have the advantage of leveraging the relation of multiple tasks. We present a skill-based multi-task RL technique on heterogeneous datasets that are generated by behavior policies of different quality. We show that our multi-task offline RL approach is robust to the mixed configurations of different-quality datasets.
  arXiv  Detail & Related papers  (2024-08-28T07:36:20Z)
• Robust Policy Learning via Offline Skill Diffusion [6.876580618014666]
  We present a novel offline skill learning framework, DuSkill. DuSkill employs a guided Diffusion model to generate versatile skills extended from the limited skills in datasets. We show that DuSkill outperforms other skill-based imitation learning and RL algorithms for several long-horizon tasks.
  arXiv  Detail & Related papers  (2024-03-01T02:00:44Z)
• Learning Options via Compression [62.55893046218824]
  We propose a new objective that combines the maximum likelihood objective with a penalty on the description length of the skills. Our objective learns skills that solve downstream tasks in fewer samples compared to skills learned from only maximizing likelihood.
  arXiv  Detail & Related papers  (2022-12-08T22:34:59Z)
• Rethinking Learning Dynamics in RL using Adversarial Networks [79.56118674435844]
  We present a learning mechanism for reinforcement learning of closely related skills parameterized via a skill embedding space. The main contribution of our work is to formulate an adversarial training regime for reinforcement learning with the help of entropy-regularized policy gradient formulation.
  arXiv  Detail & Related papers  (2022-01-27T19:51:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.