Journal ArticleDOI
Some comments of Wolfe's `away step'
J Guélat,Patrice Marcotte +1 more
TLDR
It is given a detailed proof, under slightly weaker conditions on the objective function, that a modified Frank-Wolfe algorithm based on Wolfe's ‘away step’ strategy can achieve geometric convergence, provided a strict complementarity assumption holds.Abstract:
We give a detailed proof, under slightly weaker conditions on the objective function, that a modified Frank-Wolfe algorithm based on Wolfe's ‘away step’ strategy can achieve geometric convergence, provided a strict complementarity assumption holds.read more
Citations
More filters
Proceedings ArticleDOI
Distributed frank-wolfe under pipelined stale synchronous parallelism
TL;DR: This paper proposes a model for the integration of the SSP model on a pipelined distributed processing framework, and applies SSP on a distributed version of the Frank-Wolfe algorithm, theoretically showing its sparsity bounds and convergence under SSP.
Journal ArticleDOI
Revenue optimization in energy networks involving self-scheduled demand and a smart grid
TL;DR: This paper addresses a day-ahead pricing and load balancing problem, within an environment involving self-scheduled users whose utilities are optimized via a smart grid and designs two heuristic algorithms which provide high quality solutions in moderate computation time.
Posted Content
Fast cluster detection in networks by first-order optimization
TL;DR: This paper focuses on the use of s-defective clique models for network-based cluster detection and proposes a nonlinear optimization approach that efficiently handles those models in practice and introduces an equivalent continuous formulation for the problem under analysis.
Posted Content
Generalized Self-Concordant Analysis of Frank-Wolfe algorithms
TL;DR: This paper closes the apparent gap in the literature by developing provably convergent Frank–Wolfe algorithms with standard O(1/k) convergence rate guarantees, and shows how these sublinearly convergent methods can be accelerated to yield linearly Convergent projection-free methods.
Posted Content
Novel Frank-Wolfe Methods for SVM Learning
TL;DR: A variant of the Frank-Wolfe method designed to obtain improved performance on large-scale SVM problems is presented and analyzed, based on a new way to perform away steps, a well-known strategy employed to accelerate the convergence of the basic FW method.
References
More filters