A federated optimization algorithm decouples proximal steps from communication, uses local updates and drift correction, and converges sublinearly or linearly to a bounded residual for non-convex composite losses with heterogeneous data.
Online optimization of switched LTI systems using continuous-time and hybrid accelerated gradient flows
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Non-convex composite federated learning with heterogeneous data
A federated optimization algorithm decouples proximal steps from communication, uses local updates and drift correction, and converges sublinearly or linearly to a bounded residual for non-convex composite losses with heterogeneous data.