TP4b.3

Distributed Gradient Clustering: Convergence and the Effect of Initialization

Aleksandar Armacki, Carnegie Mellon University, United States; Dragana Bajović, Faculty of Technical Sciences, University of Novi Sad, Yugoslavia; Dušan Jakovetić, Faculty of Sciences, University of Novi Sad, Yugoslavia; Soummya Kar, Carnegie Mellon University, United States
Session:
TP4b: Learning and Adaptation Lecture
Track:
Adaptive Systems, Machine Learning, and Data Analytics
Location:
Nautilus
Presentation Time:
Tue, 29 Oct, 16:20 - 16:45 PT (UTC -8)
Session Chair:
Stefan Vlaski, Imperial College London
Presentation
Discussion
Resources
No resources available.
Session TP4b
TP4b.1: On the Influence of Digraph on Decentralized Stochastic Optimization
Liyuan Liang, Peking University, China; Xinmeng Huang, University of Pennsylvania, United States; Ran Xin, ByteDance Group, United States; Kun Yuan, Peking University, China
TP4b.2: Fully Stochastic Primal-dual Decentralized Optimization for Non-convex Problems on Random Graphs
Chung-Yiu Yau, Haoming Liu, Hoi-To Wai, The Chinese University of Hong Kong, Hong Kong SAR of China
TP4b.3: Distributed Gradient Clustering: Convergence and the Effect of Initialization
Aleksandar Armacki, Carnegie Mellon University, United States; Dragana Bajović, Faculty of Technical Sciences, University of Novi Sad, Yugoslavia; Dušan Jakovetić, Faculty of Sciences, University of Novi Sad, Yugoslavia; Soummya Kar, Carnegie Mellon University, United States
TP4b.4: Decentralized Learning with Approximate Finite Time Consensus
TP4b.5: On the Inconsistency of Estimation with Uncertain Models
Cesar A Uribe, Rice University, United States; James Zachary Hare, Lance M. Kaplan, DEVCOM Army Research Laboratory, United States; Ali Jadbabaie, MIT, United States
Contacts