https://arxiv.org/abs/2412.18052 Change to arXiv's privacy policy The arXiv Privacy Policy has changed. By continuing to use arxiv.org, you are agreeing to the privacy policy. I Understand Skip to main content Cornell University We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate arxiv logo > cs > arXiv:2412.18052 [ ] Help | Advanced Search [All fields ] Search arXiv logo Cornell University Logo [ ] GO quick links * Login * Help Pages * About Computer Science > Machine Learning arXiv:2412.18052 (cs) [Submitted on 24 Dec 2024 (v1), last revised 29 Dec 2024 (this version, v2)] Title:Beyond Gradient Averaging in Parallel Optimization: Improved Robustness through Gradient Agreement Filtering Authors:Francois Chaubard, Duncan Eddy, Mykel J. Kochenderfer View a PDF of the paper titled Beyond Gradient Averaging in Parallel Optimization: Improved Robustness through Gradient Agreement Filtering, by Francois Chaubard and 2 other authors View PDF HTML (experimental) Abstract:We introduce Gradient Agreement Filtering (GAF) to improve on gradient averaging in distributed deep learning optimization. Traditional distributed data-parallel stochastic gradient descent involves averaging gradients of microbatches to calculate a macrobatch gradient that is then used to update model parameters. We find that gradients across microbatches are often orthogonal or negatively correlated, especially in late stages of training, which leads to memorization of the training set, reducing generalization. In this paper, we introduce a simple, computationally effective way to reduce gradient variance by computing the cosine distance between micro-gradients during training and filtering out conflicting updates prior to averaging. We improve validation accuracy with significantly smaller microbatch sizes. We also show this reduces memorizing noisy labels. We demonstrate the effectiveness of this technique on standard image classification benchmarks including CIFAR-100 and CIFAR-100N-Fine. We show this technique consistently outperforms validation accuracy, in some cases by up to 18.2\% compared to traditional training approaches while reducing the computation required nearly an order of magnitude because we can now rely on smaller microbatch sizes without destabilizing training. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2412.18052 [cs.LG] (or arXiv:2412.18052v2 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2412.18052 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Francois Chaubard [view email] [v1] Tue, 24 Dec 2024 00:00:11 UTC (502 KB) [v2] Sun, 29 Dec 2024 11:44:55 UTC (2,988 KB) Full-text links: Access Paper: View a PDF of the paper titled Beyond Gradient Averaging in Parallel Optimization: Improved Robustness through Gradient Agreement Filtering, by Francois Chaubard and 2 other authors * View PDF * HTML (experimental) * TeX Source * Other Formats license icon view license Current browse context: cs.LG < prev | next > new | recent | 2024-12 Change to browse by: cs cs.AI References & Citations * NASA ADS * Google Scholar * Semantic Scholar a export BibTeX citation Loading... BibTeX formatted citation x [loading... ] Data provided by: Bookmark BibSonomy logo Reddit logo (*) Bibliographic Tools Bibliographic and Citation Tools [ ] Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) [ ] Connected Papers Toggle Connected Papers (What is Connected Papers?) [ ] Litmaps Toggle Litmaps (What is Litmaps?) [ ] scite.ai Toggle scite Smart Citations (What are Smart Citations?) ( ) Code, Data, Media Code, Data and Media Associated with this Article [ ] alphaXiv Toggle alphaXiv (What is alphaXiv?) [ ] Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) [ ] DagsHub Toggle DagsHub (What is DagsHub?) [ ] GotitPub Toggle Gotit.pub (What is GotitPub?) [ ] Huggingface Toggle Hugging Face (What is Huggingface?) [ ] Links to Code Toggle Papers with Code (What is Papers with Code?) [ ] ScienceCast Toggle ScienceCast (What is ScienceCast?) ( ) Demos Demos [ ] Replicate Toggle Replicate (What is Replicate?) [ ] Spaces Toggle Hugging Face Spaces (What is Spaces?) [ ] Spaces Toggle TXYZ.AI (What is TXYZ.AI?) ( ) Related Papers Recommenders and Search Tools [ ] Link to Influence Flower Influence Flower (What are Influence Flowers?) [ ] Core recommender toggle CORE Recommender (What is CORE?) [ ] IArxiv recommender toggle IArxiv Recommender (What is IArxiv?) * Author * Venue * Institution * Topic ( ) About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) * About * Help * Click here to contact arXiv Contact * Click here to subscribe Subscribe * Copyright * Privacy Policy * Web Accessibility Assistance * arXiv Operational Status Get status notifications via email or slack