SpaFL: Communication-Efficient Federated Learning with Sparse Models and Low computational Overhead
This work addresses efficiency challenges for resource-constrained clients in federated learning, representing an incremental improvement over existing sparse methods.
The paper tackles the high communication and computation overhead in federated learning by proposing SpaFL, a framework that optimizes sparse model structures with low computational overhead, achieving improved accuracy while significantly reducing communication and computing resources compared to sparse baselines.
The large communication and computation overhead of federated learning (FL) is one of the main challenges facing its practical deployment over resource-constrained clients and systems. In this work, SpaFL: a communication-efficient FL framework is proposed to optimize sparse model structures with low computational overhead. In SpaFL, a trainable threshold is defined for each filter/neuron to prune its all connected parameters, thereby leading to structured sparsity. To optimize the pruning process itself, only thresholds are communicated between a server and clients instead of parameters, thereby learning how to prune. Further, global thresholds are used to update model parameters by extracting aggregated parameter importance. The generalization bound of SpaFL is also derived, thereby proving key insights on the relation between sparsity and performance. Experimental results show that SpaFL improves accuracy while requiring much less communication and computing resources compared to sparse baselines. The code is available at https://github.com/news-vt/SpaFL_NeruIPS_2024