Federated Hypergradient Descent

TL;DR

This paper introduces FATHOM, a one-shot, online hyperparameter optimization method for federated learning that reduces communication and computation costs by adaptively tuning hyperparameters without trial-and-error.

Contribution

FATHOM is a novel, analytical gradient-based hyperparameter optimization method for federated learning that is more communication and computationally efficient than traditional static hyperparameter tuning.

Findings

FATHOM outperforms FedAvg with static hyperparameters in communication efficiency.

FATHOM reduces overall computational costs during training.

Empirical results show improved model performance on FEMNIST and Stack Overflow datasets.

Abstract

In this work, we explore combining automatic hyperparameter tuning and optimization for federated learning (FL) in an online, one-shot procedure. We apply a principled approach on a method for adaptive client learning rate, number of local steps, and batch size. In our federated learning applications, our primary motivations are minimizing communication budget as well as local computational resources in the training pipeline. Conventionally, hyperparameter tuning methods involve at least some degree of trial-and-error, which is known to be sample inefficient. In order to address our motivations, we propose FATHOM (Federated AuTomatic Hyperparameter OptiMization) as a one-shot online procedure. We investigate the challenges and solutions of deriving analytical gradients with respect to the hyperparameters of interest. Our approach is inspired by the fact that, with the exception of local data, we have full knowledge of all components involved in our training process, and this fact can be exploited in our algorithm impactfully. We show that FATHOM is more communication efficient than Federated Averaging (FedAvg) with optimized, static valued hyperparameters, and is also more computationally efficient overall. As a communication efficient, one-shot online procedure, FATHOM solves the bottleneck of costly communication and limited local computation, by eliminating a potentially wasteful tuning process, and by optimizing the hyperparamters adaptively throughout the training procedure without trial-and-error. We show our numerical results through extensive empirical experiments with the Federated EMNIST-62 (FEMNIST) and Federated Stack Overflow (FSO) datasets, using FedJAX as our baseline framework.