FedFwd: Federated Learning without Backpropagation
This work addresses computational bottlenecks in federated learning, but it is incremental as it applies an existing BP-free method to a new context.
FedFwd tackles the problem of training efficiency in federated learning for resource-limited clients by using a backpropagation-free method, the Forward Forward algorithm, and shows competitive performance on MNIST and CIFAR-10 datasets.
In federated learning (FL), clients with limited resources can disrupt the training efficiency. A potential solution to this problem is to leverage a new learning procedure that does not rely on backpropagation (BP). We present a novel approach to FL called FedFwd that employs a recent BP-free method by Hinton (2022), namely the Forward Forward algorithm, in the local training process. FedFwd can reduce a significant amount of computations for updating parameters by performing layer-wise local updates, and therefore, there is no need to store all intermediate activation values during training. We conduct various experiments to evaluate FedFwd on standard datasets including MNIST and CIFAR-10, and show that it works competitively to other BP-dependent FL methods.