BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Collaborative learning from an optimization and game theory view
DTSTART:20230901T140000
DTEND:20230901T160000
DTSTAMP:20260407T224543Z
UID:65fcbbf3190afcf12f548e1f086869144c7a9e1adea1e442e4d61f14
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dongyang Fan\nEDIC candidacy exam\nExam president: Prof. Anne-
 Marie Kermarrec\nThesis advisor: Prof. Martin Jaggi\nCo-examiner: Prof. Da
 niel Kuhn\n\nAbstract\nIn this write-up\, we consider a decentralized lear
 ning setting where multiple agents collaborate to enhance their respective
  local objectives in terms of test performance and convergence speed. We i
 nvestigate methods for incentivizing agents to participate in collaboratio
 n and share truthful information\, optimal collaborative strategies for ag
 ents\, and the extent to which agents can derive benefits from such collab
 oration. These questions are addressed through a dual prism\, encompassing
  both optimization and game-theoretical frameworks.\n\nBackground papers\n
 - Mathieu Even\, Laurent Massoulié\, Kevin Scaman. On Sample Optimality 
 in Personalized Collaborative and Federated Learning. Neurips 2022: https
 ://proceedings.neurips.cc/paper_files/paper/2022/hash/01cea7793f3c68af2e49
 89fc66bf8fb0-Abstract-Conference.html\n- Kate Donahue\, Jon Kleinberg. M
 odel-sharing Games: Analyzing Federated Learning Under Voluntary Participa
 tion. AAAI 2021.\nhttps://ojs.aaai.org/index.php/AAAI/article/view/16669\n
 - Florian E. Dorner and Nikola Konstantinov and Georgi Pashaliev and Mart
 in Vechev. Incentivizing Honesty among Competitors in Collaborative Learn
 ing and Optimization. Arxiv 2023\nhttps://arxiv.org/abs/2305.16272\n\n 
LOCATION:BC 329 https://plan.epfl.ch/?room==BC%20329
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR