COALA: A Practical and Vision-Centric Federated Learning Platform

TL;DR

COALA is a comprehensive federated learning platform tailored for practical vision tasks, supporting diverse data, task, and model configurations, and providing benchmarks for real-world scenarios.

Contribution

It introduces a versatile FL platform with extensive support for vision tasks, data types, and model configurations, along with systematic benchmarks for practical FL scenarios.

Findings

Extended FL support to 15 vision tasks including detection and segmentation.

Benchmarked semi-supervised, unsupervised, and continual FL scenarios.

Identified opportunities for advancing federated learning techniques.

Abstract

We present COALA, a vision-centric Federated Learning (FL) platform, and a suite of benchmarks for practical FL scenarios, which we categorize into three levels: task, data, and model. At the task level, COALA extends support from simple classification to 15 computer vision tasks, including object detection, segmentation, pose estimation, and more. It also facilitates federated multiple-task learning, allowing clients to tackle multiple tasks simultaneously. At the data level, COALA goes beyond supervised FL to benchmark both semi-supervised FL and unsupervised FL. It also benchmarks feature distribution shifts other than commonly considered label distribution shifts. In addition to dealing with static data, it supports federated continual learning for continuously changing data in real-world scenarios. At the model level, COALA benchmarks FL with split models and different models in different clients. COALA platform offers three degrees of customization for these practical FL scenarios, including configuration customization, components customization, and workflow customization. We conduct systematic benchmarking experiments for the practical FL scenarios and highlight potential opportunities for further advancements in FL. Codes are open sourced at https://github.com/SonyResearch/COALA.