FLoCoRA: Federated learning compression with low-rank adaptation
This work addresses communication efficiency for federated learning practitioners, presenting an incremental improvement by adapting existing LoRA methods to a new context.
The paper tackles the problem of high communication costs in federated learning by applying Low-Rank Adaptation (LoRA) to train small-vision models from scratch, achieving a 4.8x reduction in communication with less than 1% accuracy degradation on CIFAR-10 with ResNet-8, and extending it with quantization to achieve an 18.6x reduction with similar accuracy loss on ResNet-18.
Low-Rank Adaptation (LoRA) methods have gained popularity in efficient parameter fine-tuning of models containing hundreds of billions of parameters. In this work, instead, we demonstrate the application of LoRA methods to train small-vision models in Federated Learning (FL) from scratch. We first propose an aggregation-agnostic method to integrate LoRA within FL, named FLoCoRA, showing that the method is capable of reducing communication costs by 4.8 times, while having less than 1% accuracy degradation, for a CIFAR-10 classification task with a ResNet-8. Next, we show that the same method can be extended with an affine quantization scheme, dividing the communication cost by 18.6 times, while comparing it with the standard method, with still less than 1% of accuracy loss, tested with on a ResNet-18 model. Our formulation represents a strong baseline for message size reduction, even when compared to conventional model compression works, while also reducing the training memory requirements due to the low-rank adaptation.