CrossFL-2022: Workshop on Cross-Community Federated Learning: Algorithms, Systems and Co-designs

Recordings of invited talks given at CrossFL-2022: Workshop on Cross-Community Federated Learning: Algorithms, Systems and Co-designs, co-located with MLSys2022 (Hybrid workshop in Santa Clara Convention Center, CA and virtual) on September 1, 2022.

Workshop webpage: https://crossfl2022.github.io/

Contents:
1. Opening Remarks by Tianyi Chen (RPI)
2. Keynote #1 - ProxSkip: Yes! Local Gradient Steps Provably Lead to Communication Acceleration! Finally! by Peter Richtarik (KAUST)
3. Keynote #2 - Three Daunting Challenges of Federated Learning: Privacy Leakage, Label Deficiency, and Resource Constraints by Salman Avestimehr (USC)
4. Keynote #3 - Federated Learning for EdgeAI: New Ideas and Opportunities for Progress by Radu Marculescu (UT Austin)
5. Keynote #4 - Model Based Deep Learning with Applications to Federated Learning by Yonina Eldar (Weizmann)
6. Live Demo Session on FedML by Chaoyang He (FedML)
7. Keynote #5 - Scalable, Heterogeneity-Aware and Trustworthy Federated Learning by Yiran Chen (Duke)
8. Keynote #6 - On Lower Bounds of Distributed Learning with Communication Compression by Wotao Yin (Alibaba Damo)

Видео CrossFL-2022: Workshop on Cross-Community Federated Learning: Algorithms, Systems and Co-designs канала TrustworthyAI