English

Federated Learning Under Temporal Drift -- Mitigating Catastrophic Forgetting via Experience Replay

Machine Learning 2026-01-21 v1 Distributed, Parallel, and Cluster Computing

Abstract

Federated Learning struggles under temporal concept drift where client data distributions shift over time. We demonstrate that standard FedAvg suffers catastrophic forgetting under seasonal drift on Fashion-MNIST, with accuracy dropping from 74% to 28%. We propose client-side experience replay, where each client maintains a small buffer of past samples mixed with current data during local training. This simple approach requires no changes to server aggregation. Experiments show that a 50-sample-per-class buffer restores performance to 78-82%, effectively preventing forgetting. Our ablation study reveals a clear memory-accuracy trade-off as buffer size increases.

Keywords

Cite

@article{arxiv.2601.13456,
  title  = {Federated Learning Under Temporal Drift -- Mitigating Catastrophic Forgetting via Experience Replay},
  author = {Sahasra Kokkula and Daniel David and Aaditya Baruah},
  journal= {arXiv preprint arXiv:2601.13456},
  year   = {2026}
}

Comments

8 pages, 5 figures. Course project for Neural Networks & Deep Learning COMSW4776 course at Columbia University

R2 v1 2026-07-01T09:11:33.221Z