Averaging Rate Scheduler for Decentralized Learning on Heterogeneous Data

March 05, 2024 ยท Declared Dead ยท ๐Ÿ› Tiny Papers @ ICLR

๐Ÿ‘ป CAUSE OF DEATH: Ghosted
No code link whatsoever

"No code URL or promise found in abstract"

Evidence collected by the PWNC Scanner

Authors Sai Aparna Aketi, Sakshi Choudhary, Kaushik Roy arXiv ID 2403.03292 Category cs.LG: Machine Learning Cross-listed cs.DC Citations 2 Venue Tiny Papers @ ICLR Last Checked 5 months ago
Abstract
State-of-the-art decentralized learning algorithms typically require the data distribution to be Independent and Identically Distributed (IID). However, in practical scenarios, the data distribution across the agents can have significant heterogeneity. In this work, we propose averaging rate scheduling as a simple yet effective way to reduce the impact of heterogeneity in decentralized learning. Our experiments illustrate the superiority of the proposed method (~3% improvement in test accuracy) compared to the conventional approach of employing a constant averaging rate.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Machine Learning

Died the same way โ€” ๐Ÿ‘ป Ghosted