Federated Multi-Task Learning on Non-IID Data Silos: An Experimental Study
February 20, 2024 ยท Entered Twilight ยท ๐ International Conference on Multimedia Retrieval
Repo contents: .gitignore, README.md, aggregate_all.py, aggregate_md.py, configs, convert.py, datasets, evaluation, losses.py, main.py, main_md.py, models, pick_figs.py, test.py, utils.py
Authors
Yuwen Yang, Yuxiang Lu, Suizhi Huang, Shalayiding Sirejiding, Hongtao Lu, Yue Ding
arXiv ID
2402.12876
Category
cs.LG: Machine Learning
Cross-listed
cs.CR,
cs.DC
Citations
9
Venue
International Conference on Multimedia Retrieval
Repository
https://github.com/youngfish42/FMTL-Benchmark
โญ 9
Last Checked
6 months ago
Abstract
The innovative Federated Multi-Task Learning (FMTL) approach consolidates the benefits of Federated Learning (FL) and Multi-Task Learning (MTL), enabling collaborative model training on multi-task learning datasets. However, a comprehensive evaluation method, integrating the unique features of both FL and MTL, is currently absent in the field. This paper fills this void by introducing a novel framework, FMTL-Bench, for systematic evaluation of the FMTL paradigm. This benchmark covers various aspects at the data, model, and optimization algorithm levels, and comprises seven sets of comparative experiments, encapsulating a wide array of non-independent and identically distributed (Non-IID) data partitioning scenarios. We propose a systematic process for comparing baselines of diverse indicators and conduct a case study on communication expenditure, time, and energy consumption. Through our exhaustive experiments, we aim to provide valuable insights into the strengths and limitations of existing baseline methods, contributing to the ongoing discourse on optimal FMTL application in practical scenarios. The source code can be found on https://github.com/youngfish42/FMTL-Benchmark .
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
๐ Similar Papers
In the same crypt โ Machine Learning
๐ฎ
๐ฎ
The Ethereal
๐ฎ
๐ฎ
The Ethereal
Continuous control with deep reinforcement learning
๐
๐
Old Age
Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
๐
๐
Old Age
Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
๐
๐
Old Age
SGDR: Stochastic Gradient Descent with Warm Restarts
๐ฎ
๐ฎ
The Ethereal