Scholarlink

This paper introduces ACoaL (Asynchronous Consensus-based with coalitions Learning), a novel Federated Learning algorithm that enhances the learning process by forming coalitions within a MAS. ACoaL builds upon Co-Learning, focusing on intra-coalition communication to strengthen learning. The algorithm leverages SPADE framework for agent communication and coordination. The paper presents a case study on fruit classification to demonstrate ACoaL’s effectiveness, highlighting its potential for distributed learning tasks.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Introducing Coalitions to Improve the Performance of Federated Learning Consensus-Based Algorithms

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Francisco Enguix,
  • J. A. Rincon,
  • C. Carrascosa

摘要

This paper introduces ACoaL (Asynchronous Consensus-based with coalitions Learning), a novel Federated Learning algorithm that enhances the learning process by forming coalitions within a MAS. ACoaL builds upon Co-Learning, focusing on intra-coalition communication to strengthen learning. The algorithm leverages SPADE framework for agent communication and coordination. The paper presents a case study on fruit classification to demonstrate ACoaL’s effectiveness, highlighting its potential for distributed learning tasks.