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.
@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