Achieving acceleration in distributed optimization via direct discretization of the heavy-ball ODE

Achieving acceleration in distributed optimization via direct discretization of the heavy-ball ODE

© 2019 American Automatic Control Council. We develop a distributed algorithm for convex Empirical Risk Minimization, the problem of minimizing large but finite sum of convex functions over networks. The proposed algorithm is derived from directly discretizing the second-order heavy-ball differentia...

Πλήρης περιγραφή

Λεπτομέρειες βιβλιογραφικής εγγραφής
Κύριοι συγγραφείς: Zhang, J, Uribe, CA, Mokhtari, A, Jadbabaie, A
Άλλοι συγγραφείς: Massachusetts Institute of Technology. Department of Civil and Environmental Engineering
Μορφή: Άρθρο
Γλώσσα:English
Έκδοση: IEEE 2023
Διαθέσιμο Online:https://hdl.handle.net/1721.1/148596

Διαδίκτυο

https://hdl.handle.net/1721.1/148596

Παρόμοια τεκμήρια