FLBench: A Benchmark Suite for Federated Learning

TL;DR

FLBench is a comprehensive, domain-diverse benchmark suite designed to evaluate federated learning systems and algorithms in realistic scenarios, addressing limitations of previous simulation-based benchmarks.

Contribution

The paper introduces FLBench, a novel federated learning benchmark suite with real-world domain configurations and an automated deployment tool, enhancing evaluation fidelity.

Findings

FLBench covers medical, financial, and AIoT domains.

It evaluates communication, privacy, data heterogeneity, and cooperation strategies.

Open sourced and actively evolving platform.

Abstract

Federated learning is a new machine learning paradigm. The goal is to build a machine learning model from the data sets distributed on multiple devices so-called an isolated data island, while keeping their data secure and private. Most existing federated learning benchmarks work manually splits commonly used public datasets into partitions to simulate real world isolated data island scenarios. Still, this simulation fails to capture real world isolated data island intrinsic characteristics. This paper presents a federated learning (FL) benchmark suite named FLBench. FLBench contains three domains: medical, financial, and AIoT. By configuring various domains, FLBench is qualified to evaluate federated learning systems and algorithms essential aspects, like communication, scenario transformation, privacy-preserving, data distribution heterogeneity, and cooperation strategy. Hence, it becomes a promising platform for developing novel federated learning algorithms. Currently, FLBench is open sourced and in fast evolution. We package it as an automated deployment tool. The benchmark suite is available from https://www.benchcouncil.org/flbench.html.