Loading...
conference paper
Distributed Meta-Learning with Networked Agents
January 1, 2021
29Th European Signal Processing Conference (Eusipco 2021)
Meta-learning aims to improve efficiency of learning new tasks by exploiting the inductive biases obtained from related tasks. Previous works consider centralized or federated architectures that rely on central processors, whereas, in this paper, we propose a decentralized meta-learning scheme where the data and the computations are distributed across a network of agents. We provide convergence results for non-convex environments and illustrate the theoretical findings with experiments.
Type
conference paper
Web of Science ID
WOS:000764066600271
Authors
Publication date
2021-01-01
Published in
29Th European Signal Processing Conference (Eusipco 2021)
ISBN of the book
978-9-0827-9706-0
Publisher place
Kessariani
Series title/Series vol.
European Signal Processing Conference
Start page
1361
End page
1365
Peer reviewed
REVIEWED
EPFL units
Event name | Event place | Event date |
ELECTR NETWORK | Aug 23-27, 2021 | |
Available on Infoscience
April 25, 2022
Use this identifier to reference this record