SS14.4
Smoothing ADMM for Non-convex and Non-smooth Hierarchical Federated Learning
Reza Mirzaeifard, Stefan Werner, Norwegian University of Science and Technology, Norway
Session:
SS14: Multi-Objective Optimization and Learning – Algorithms and Applications Lecture
Track:
Special Sessions
Location:
Teatro del Sole
Presentation Time:
Thu, 11 Sep, 10:00 - 10:20 Italy Time (UTC +2)
Session Co-Chairs:
Stefan Vlaski, and Sangwoo Park,
Presentation
Discussion
Resources
No resources available.
Session SS14
SS14.1: ENSURING RELIABILITY VIA HYPERPARAMETER SELECTION: REVIEW AND ADVANCES
Amirmohammad Farzaneh, Osvaldo Simeone, King's College London, United Kingdom
SS14.2: D-SOBA: A Single-Loop Decentralized Bilevel Algorithm with Transient Complexity Analysis
Boao Kong, Shuchen Zhu, Peking University, China; Songtao Lu, The Chinese University of Hong Kong, China; Xinmeng Huang, University of Pennsylvania, United States; Kun Yuan, Peking University, China
SS14.3: Multivariate Belief Formation for Distributed Multitask Learning
Malek Khammassi, École Polytechnique Fédérale de Lausanne, Switzerland; Vincenzo Matta, University of Salerno, Italy; Ali H. Sayed, École Polytechnique Fédérale de Lausanne, Switzerland
SS14.4: Smoothing ADMM for Non-convex and Non-smooth Hierarchical Federated Learning
Reza Mirzaeifard, Stefan Werner, Norwegian University of Science and Technology, Norway
SS14.5: Improved Analysis of Penalty-Based Methods for Bilevel Optimization with Coupled Constraints
Liuyuan Jiang, University of Rochester, United States; Quan Xiao, Tianyi Chen, Cornell Tech, Cornell University, United States