Concept-aware clustering for decentralized deep learning under temporal shift

Edvin Listo Zec, Emilie Klefbom, Marcus Toftås, Martin Johan Willbo, Olof Mogren

Keywords: decentralized learning, deep learning, shift

TL;DR: We study temporal distributional shifts in decentralized deep learning

Abstract: Decentralized deep learning requires dealing with non-iid data across clients, which may also change over time due to temporal shifts. While non-iid data has been extensively studied in distributed settings, temporal shifts have received no attention. To the best of our knowledge, we are first with tackling the novel and challenging problem of decentralized learning with non-iid and dynamic data. We propose a novel algorithm that can automatically discover and adapt to the evolving concepts in the network, without any prior knowledge or estimation of the number of concepts. We evaluate our algorithm on standard benchmark datasets and demonstrate that it outperforms previous methods for decentralized learning.

Submission Number: 22