Tram-FL: Reducing Communication and Computation Costs through Sequential Model Circulation in Decentralized Federated Learning

Key points

• Conventional decentralized federated learning (DFL) often focuses on clients, with each client maintaining a model copy, performing updates individually, and undertaking model exchange and integration.1
• We address the training scheduling problem in model circulation-based training, specifically determining which nodes should update the model and the number of updates to perform.1
• This is approached by considering the model's circulation route and update iteration allocation, for which we propose simple yet effective methods.1
• Experimental results show that the proposed algorithm, even with non-IID data, converges to a global model with reduced communication and computation.1

Sources (1)

• [1]Tram-FL: Reducing Communication and Computation Costs through Sequential Model Circulation in Decentralized Federated Learning
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 07:01 AM

  To this end, we propose Tram-FL (Traveling Model Training Mechanism for Decentralized Federated Learning), a mechanism designed to efficiently address these challenges.

  Conventional decentralized federated learning (DFL) often focuses on clients, with each client maintaining a model copy, performing updates individually, and undertaking model exchange and integration.

Extractive summary: sentences quoted from the sources.