Logo image
Cache-Enabled Federated Learning Systems
Conference proceeding   Open access

Cache-Enabled Federated Learning Systems

Yuezhou Liu, Lili Su, Carlee Joe-Wong, Stratis Ioannidis, Edmund Yeh and Marie Siew
Proceedings of the Twenty-fourth International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing, pp.1-11
ACM Conferences
MobiHoc '23: Twenty-fourth International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing
23/10/2023

Abstract

Computing methodologies -- Distributed computing methodologies -- Distributed algorithms Computing methodologies -- Machine learning Computing methodologies -- Parallel computing methodologies -- Parallel algorithms Networks -- Network services
Federated learning (FL) is a distributed paradigm for collaboratively learning models without having clients disclose their private data. One natural and practically relevant metric to measure the efficiency of FL algorithms is the total wall-clock training time, which can be quantified by the product of the average time needed for a single iteration and the number of iterations for convergence. In this work, we focus on improving FL efficiency with respect to this metric through caching. Specifically, instead of having all clients download the latest global model from a parameter server, we select a subset of clients to access, with a smaller delay, a somewhat stale global model stored in caches. We propose CacheFL - a cache-enabled variant of FedAvg, and provide theoretical convergence guarantees in the general setting where the local data is imbalanced and heterogeneous. Armed with this result, we determine the caching strategies that minimize total wall-clock training time at a given convergence threshold for both stochastic and deterministic communication/computation delays. Through numerical experiments on real data traces, we show the advantage of our proposed scheme against several baselines, over both synthetic and real-world datasets.
url
https://doi.org/10.1145/3565287.3610264View
Published (Version of record) Open

Metrics

1 Record Views

Details

Logo image