FedZeN: Towards superlinear zeroth-order federated learning via incremental Hessian estimation

TL;DR

FedZeN introduces a novel federated zeroth-order optimization algorithm that estimates the Hessian incrementally, achieving superlinear convergence in convex settings while maintaining communication efficiency and privacy.

Contribution

This work is the first to develop a federated zeroth-order method with incremental Hessian estimation for superlinear convergence in convex optimization.

Findings

Achieves local quadratic convergence with high probability.

Demonstrates global linear convergence up to zeroth-order precision.

Outperforms existing federated zeroth-order methods in simulations.

Abstract

Federated learning is a distributed learning framework that allows a set of clients to collaboratively train a model under the orchestration of a central server, without sharing raw data samples. Although in many practical scenarios the derivatives of the objective function are not available, only few works have considered the federated zeroth-order setting, in which functions can only be accessed through a budgeted number of point evaluations. In this work we focus on convex optimization and design the first federated zeroth-order algorithm to estimate the curvature of the global objective, with the purpose of achieving superlinear convergence. We take an incremental Hessian estimator whose error norm converges linearly, and we adapt it to the federated zeroth-order setting, sampling the random search directions from the Stiefel manifold for improved performance. In particular, both the gradient and Hessian estimators are built at the central server in a communication-efficient and privacy-preserving way by leveraging synchronized pseudo-random number generators. We provide a theoretical analysis of our algorithm, named FedZeN, proving local quadratic convergence with high probability and global linear convergence up to zeroth-order precision. Numerical simulations confirm the superlinear convergence rate and show that our algorithm outperforms the federated zeroth-order methods available in the literature.