University of Illinois Chicago
Browse

Flag Aggregator: Scalable Distributed Training under Failures and Augmented Losses using Convex Optimization

Download (1.82 MB)
conference contribution
posted on 2024-07-13, 22:00 authored by Hamidreza Almasi, Harsh Mishra, Balajee VamananBalajee Vamanan, Sathya N Ravi
No description supplied

Funding

Distributed ML Training with Large-scale Accelerator Datasets for Automated Performance Optimization | Funder: Discovery Partner Institute (DPI)

Optimizing Distributed Training for Large and Noisy Data | Funder: CAHSI-Google Institutional Research Program (NSF, Google)

History

Citation

Almasi, H., Mishra, H., Vamanan, B.Ravi, S. N. (2024, May). Flag Aggregator: Scalable Distributed Training under Failures and Augmented Losses using Convex Optimization. International Conference on Learning Representations (ICLR), Vienna, Austria.

Usage metrics

    Categories

    No categories selected

    Keywords

    Uncategorised value

    Licence

    In Copyright

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC