Policy Search, Retrieval, and Composition via Task Similarity in Collaborative Agentic Systems

• URL: http://arxiv.org/abs/2506.05577v2
• Date: Mon, 18 Aug 2025 13:42:36 GMT
• Title: Policy Search, Retrieval, and Composition via Task Similarity in Collaborative Agentic Systems
• Authors: Saptarshi Nath, Christos Peridis, Eseoghene Benjamin, Xinran Liu, Soheil Kolouri, Peter Kinnell, Zexin Li, Cong Liu, Shirin Dora, Andrea Soltoggio,
• Abstract summary: Agentic AI aims to create systems that set their own goals, adapt proactively to change, and refine behavior through continuous experience.<n>Recent advances suggest that, when facing multiple and unforeseen tasks, agents could benefit from sharing machine-learned knowledge and reuse policies that have already been fully or partially learned by other agents.<n>This study explores how an agent decides what knowledge to select, from whom, and when and how to integrate it in its own policy in order to accelerate its own learning.
• Score: 12.471774408499817
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Agentic AI aims to create systems that set their own goals, adapt proactively to change, and refine behavior through continuous experience. Recent advances suggest that, when facing multiple and unforeseen tasks, agents could benefit from sharing machine-learned knowledge and reuse policies that have already been fully or partially learned by other agents. However, how to query, select, and retrieve policies from a pool of agents, and how to integrate such policies remains a largely unexplored area. This study explores how an agent decides what knowledge to select, from whom, and when and how to integrate it in its own policy in order to accelerate its own learning. The proposed algorithm, \emph{Modular Sharing and Composition in Collective Learning} (MOSAIC), improves learning in agentic collectives by combining (1) knowledge selection using performance signals and cosine similarity on Wasserstein task embeddings, (2) modular and transferable neural representations via masks, and (3) policy integration, composition and fine-tuning. MOSAIC outperforms isolated learners and global sharing approaches in both learning speed and overall performance, and in some cases solves tasks that isolated agents cannot. The results also demonstrate that selective, goal-driven reuse leads to less susceptibility to task interference. We also observe the emergence of self-organization, where agents solving simpler tasks accelerate the learning of harder ones through shared knowledge.

Related papers

• Cross-Task Experiential Learning on LLM-based Multi-Agent Collaboration [63.90193684394165]
  We introduce multi-agent cross-task experiential learning (MAEL), a novel framework that endows LLM-driven agents with explicit cross-task learning and experience accumulation.<n>During the experiential learning phase, we quantify the quality for each step in the task-solving workflow and store the resulting rewards.<n>During inference, agents retrieve high-reward, task-relevant experiences as few-shot examples to enhance the effectiveness of each reasoning step.
  arXiv  Detail & Related papers  (2025-05-29T07:24:37Z)
• Contextual Knowledge Sharing in Multi-Agent Reinforcement Learning with Decentralized Communication and Coordination [0.9776703963093367]
  Multi-Agent Reinforcement Learning (Dec-MARL) has emerged as a pivotal approach for addressing complex tasks in dynamic environments.<n>This paper presents a novel Dec-MARL framework that integrates peer-to-peer communication and coordination, incorporating goal-awareness and time-awareness into the agents' knowledge-sharing processes.
  arXiv  Detail & Related papers  (2025-01-26T22:49:50Z)
• Capability-Aware Shared Hypernetworks for Flexible Heterogeneous Multi-Robot Coordination [2.6590401523087634]
  We propose Capability-Aware Shared Hypernetworks (CASH) for multi-robot teams.<n>CASH is a soft weight sharing architecture that uses hypernetworks to efficiently learn a flexible shared policy.<n>We show that CASH consistently outperforms baseline architectures in terms of performance and sample efficiency during both training and zero-shot generalization.
  arXiv  Detail & Related papers  (2025-01-10T15:39:39Z)
• Collaborative Gym: A Framework for Enabling and Evaluating Human-Agent Collaboration [51.452664740963066]
  Collaborative Gym is a framework enabling asynchronous, tripartite interaction among agents, humans, and task environments.<n>We instantiate Co-Gym with three representative tasks in both simulated and real-world conditions.<n>Our findings reveal that collaborative agents consistently outperform their fully autonomous counterparts in task performance.
  arXiv  Detail & Related papers  (2024-12-20T09:21:15Z)
• Tacit Learning with Adaptive Information Selection for Cooperative Multi-Agent Reinforcement Learning [13.918498667158119]
  We introduce a novel cooperative MARL framework based on information selection and tacit learning.<n>We integrate gating and selection mechanisms, allowing agents to adaptively filter information based on environmental changes.<n>Experiments on popular MARL benchmarks show that our framework can be seamlessly integrated with state-of-the-art algorithms.
  arXiv  Detail & Related papers  (2024-12-20T07:55:59Z)
• From Novice to Expert: LLM Agent Policy Optimization via Step-wise Reinforcement Learning [62.54484062185869]
  We introduce StepAgent, which utilizes step-wise reward to optimize the agent's reinforcement learning process.<n>We propose implicit-reward and inverse reinforcement learning techniques to facilitate agent reflection and policy adjustment.
  arXiv  Detail & Related papers  (2024-11-06T10:35:11Z)
• Multi-agent cooperation through learning-aware policy gradients [53.63948041506278]
  Self-interested individuals often fail to cooperate, posing a fundamental challenge for multi-agent learning.<n>We present the first unbiased, higher-derivative-free policy gradient algorithm for learning-aware reinforcement learning.<n>We derive from the iterated prisoner's dilemma a novel explanation for how and when cooperation arises among self-interested learning-aware agents.
  arXiv  Detail & Related papers  (2024-10-24T10:48:42Z)
• I Know How: Combining Prior Policies to Solve New Tasks [17.214443593424498]
  Multi-Task Reinforcement Learning aims at developing agents that are able to continually evolve and adapt to new scenarios. Learning from scratch for each new task is not a viable or sustainable option. We propose a new framework, I Know How, which provides a common formalization.
  arXiv  Detail & Related papers  (2024-06-14T08:44:51Z)
• Distributed Continual Learning [12.18012293738896]
  We introduce a mathematical framework capturing the essential aspects of distributed continual learning.<n>We identify three modes of information exchange: data instances, full model parameters, and modular (partial) model parameters.<n>Our findings reveal three key insights: sharing parameters is more efficient than sharing data as tasks become more complex.
  arXiv  Detail & Related papers  (2024-05-23T21:24:26Z)
• Decentralized and Lifelong-Adaptive Multi-Agent Collaborative Learning [57.652899266553035]
  Decentralized and lifelong-adaptive multi-agent collaborative learning aims to enhance collaboration among multiple agents without a central server. We propose DeLAMA, a decentralized multi-agent lifelong collaborative learning algorithm with dynamic collaboration graphs.
  arXiv  Detail & Related papers  (2024-03-11T09:21:11Z)
• Learning in Cooperative Multiagent Systems Using Cognitive and Machine Models [1.0742675209112622]
  Multi-Agent Systems (MAS) are critical for many applications requiring collaboration and coordination with humans. One major challenge is the simultaneous learning and interaction of independent agents in dynamic environments. We propose three variants of Multi-Agent IBL models (MAIBL) We demonstrate that the MAIBL models exhibit faster learning and achieve better coordination in a dynamic CMOTP task with various settings of rewards compared to current MADRL models.
  arXiv  Detail & Related papers  (2023-08-18T00:39:06Z)
• Autonomous Open-Ended Learning of Tasks with Non-Stationary Interdependencies [64.0476282000118]
  Intrinsic motivations have proven to generate a task-agnostic signal to properly allocate the training time amongst goals. While the majority of works in the field of intrinsically motivated open-ended learning focus on scenarios where goals are independent from each other, only few of them studied the autonomous acquisition of interdependent tasks. In particular, we first deepen the analysis of a previous system, showing the importance of incorporating information about the relationships between tasks at a higher level of the architecture. Then we introduce H-GRAIL, a new system that extends the previous one by adding a new learning layer to store the autonomously acquired sequences
  arXiv  Detail & Related papers  (2022-05-16T10:43:01Z)
• Collaborative Training of Heterogeneous Reinforcement Learning Agents in Environments with Sparse Rewards: What and When to Share? [7.489793155793319]
  This work focuses on combining information obtained through intrinsic motivation with the aim of having a more efficient exploration and faster learning. Our results reveal different ways in which a collaborative framework with little additional computational cost can outperform an independent learning process without knowledge sharing.
  arXiv  Detail & Related papers  (2022-02-24T16:15:51Z)
• Constructing a Good Behavior Basis for Transfer using Generalized Policy Updates [63.58053355357644]
  We study the problem of learning a good set of policies, so that when combined together, they can solve a wide variety of unseen reinforcement learning tasks. We show theoretically that having access to a specific set of diverse policies, which we call a set of independent policies, can allow for instantaneously achieving high-level performance.
  arXiv  Detail & Related papers  (2021-12-30T12:20:46Z)
• Learning When and What to Ask: a Hierarchical Reinforcement Learning Framework [17.017688226277834]
  We formulate a hierarchical reinforcement learning framework for learning to decide when to request additional information from humans. Results on a simulated human-assisted navigation problem demonstrate the effectiveness of our framework.
  arXiv  Detail & Related papers  (2021-10-14T01:30:36Z)
• Behaviour-conditioned policies for cooperative reinforcement learning tasks [41.74498230885008]
  In various real-world tasks, an agent needs to cooperate with unknown partner agent types. Deep reinforcement learning models can be trained to deliver the required functionality but are known to suffer from sample inefficiency and slow learning. We suggest a method, where we synthetically produce populations of agents with different behavioural patterns together with ground truth data of their behaviour. We additionally suggest an agent architecture, which can efficiently use the generated data and gain the meta-learning capability.
  arXiv  Detail & Related papers  (2021-10-04T09:16:41Z)
• Multi-Agent Imitation Learning with Copulas [102.27052968901894]
  Multi-agent imitation learning aims to train multiple agents to perform tasks from demonstrations by learning a mapping between observations and actions. In this paper, we propose to use copula, a powerful statistical tool for capturing dependence among random variables, to explicitly model the correlation and coordination in multi-agent systems. Our proposed model is able to separately learn marginals that capture the local behavioral patterns of each individual agent, as well as a copula function that solely and fully captures the dependence structure among agents.
  arXiv  Detail & Related papers  (2021-07-10T03:49:41Z)
• Latent Skill Planning for Exploration and Transfer [49.25525932162891]
  In this paper, we investigate how these two approaches can be integrated into a single reinforcement learning agent. We leverage the idea of partial amortization for fast adaptation at test time. We demonstrate the benefits of our design decisions across a suite of challenging locomotion tasks.
  arXiv  Detail & Related papers  (2020-11-27T18:40:03Z)
• UneVEn: Universal Value Exploration for Multi-Agent Reinforcement Learning [53.73686229912562]
  We propose a novel MARL approach called Universal Value Exploration (UneVEn) UneVEn learns a set of related tasks simultaneously with a linear decomposition of universal successor features. Empirical results on a set of exploration games, challenging cooperative predator-prey tasks requiring significant coordination among agents, and StarCraft II micromanagement benchmarks show that UneVEn can solve tasks where other state-of-the-art MARL methods fail.
  arXiv  Detail & Related papers  (2020-10-06T19:08:47Z)

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.