FedHPO-B: A Benchmark Suite for Federated Hyperparameter Optimization

TL;DR

FedHPO-B is a new benchmark suite designed specifically for hyperparameter optimization in federated learning, addressing the unique challenges and enabling fair comparison of HPO methods in this setting.

Contribution

The paper introduces FedHPO-B, the first comprehensive benchmark suite tailored for federated hyperparameter optimization, including diverse tasks and evaluation protocols.

Findings

Existing HPO benchmarks are inadequate for FL.

FedHPO-B facilitates fair comparison of HPO methods in FL.

Extensive experiments benchmark several HPO algorithms.

Abstract

Hyperparameter optimization (HPO) is crucial for machine learning algorithms to achieve satisfactory performance, whose progress has been boosted by related benchmarks. Nonetheless, existing efforts in benchmarking all focus on HPO for traditional centralized learning while ignoring federated learning (FL), a promising paradigm for collaboratively learning models from dispersed data. In this paper, we first identify some uniqueness of HPO for FL algorithms from various aspects. Due to this uniqueness, existing HPO benchmarks no longer satisfy the need to compare HPO methods in the FL setting. To facilitate the research of HPO in the FL setting, we propose and implement a benchmark suite FedHPO-B that incorporates comprehensive FL tasks, enables efficient function evaluations, and eases continuing extensions. We also conduct extensive experiments based on FedHPO-B to benchmark a few HPO methods. We open-source FedHPO-B at https://github.com/alibaba/FederatedScope/tree/master/benchmark/FedHPOB.