Federated Learning with Client Subsampling, Data Heterogeneity, and Unbounded Smoothness: A New Algorithm and Lower Bounds Michael Crawshaw

Feb-8-2026, 04:16:08 GMT–Neural Information Processing Systems

Lastly, we provide an experimental evaluation of EPISODE++ when training RNNs on federated text classification tasks, demonstrating that EPISODE++ outperforms strong baselines in FL.

bl 1, machine learning, natural language, (16 more...)

Neural Information Processing Systems

Feb-8-2026, 04:16:08 GMT

Conferences PDF

Country:
- Asia > China
  - Shanghai > Shanghai (0.04)
- North America > United States
  - Massachusetts > Middlesex County
    - Cambridge (0.04)
  - Virginia (0.04)

Technology:
- Information Technology > Artificial Intelligence
  - Machine Learning > Neural Networks
    - Deep Learning (0.93)
  - Natural Language (1.00)

Duplicate Docs Excel Report

Title
Federated Learning with Client Subsampling, Data Heterogeneity, and Unbounded Smoothness: A New Algorithm and Lower Bounds

Similar Docs Excel Report more

Title	Similarity	Source
None found