Modern large-scale learning increasingly relies on decentralized systems to scale computation, eliminate central points of failure, and enable training in communication-constrained or privacy-sensitive environments. Yet, despite their promise, decentralized learning algorithms face fundamental scalability limitations due to delayed consensus and model divergence.

The Department of Computer Science at the University of Salzburg warmly invites you to the talk by Assoc. Prof. Kfir Y. Levy on

📅 Tuesday, July 15, 2025

🕒 11:00 AM

📍 Lecture Hall T01, Jakob-Haringer-Straße 2 (JAK2, ground floor, room 0.01)

Host: Univ.-Prof. Sebastian Forster

Guest Lecture_Kfir Y. Levy 15.07.2025