Online Decentralized Frank-Wolfe: From theoretical bound to applications in smart-building
Machine Learning
2022-08-02 v1 Data Structures and Algorithms
Optimization and Control
Abstract
The design of decentralized learning algorithms is important in the fast-growing world in which data are distributed over participants with limited local computation resources and communication. In this direction, we propose an online algorithm minimizing non-convex loss functions aggregated from individual data/models distributed over a network. We provide the theoretical performance guarantee of our algorithm and demonstrate its utility on a real life smart building.
Keywords
Cite
@article{arxiv.2208.00522,
title = {Online Decentralized Frank-Wolfe: From theoretical bound to applications in smart-building},
author = {Angan Mitra and Nguyen Kim Thang and Tuan-Anh Nguyen and Denis Trystram and Paul Youssef},
journal= {arXiv preprint arXiv:2208.00522},
year = {2022}
}