KAUST's Richtárik releases BroxGD paper approximating constraint-ball intersections via Frank-Wolfe
peter_richtarik · x · 2026-11-05
KAUST professor Peter Richtárik announced BroxGD, a new optimization paper co-authored with his M.Sc. student Zhirayr Tovmasyan. The thread includes a visualization showing how the intersection of the constraint set and a ball in BroxGD is approximated via Frank-Wolfe.
The group focuses on optimization for large-scale and distributed learning, with related work including error feedback for Muon and variance-reduction methods for parallel stochastic optimization.
Related event: KAUST's BroxGD Paper Clarifies Frank-Wolfe as Projection-Free Approximation(3 posts)→
More from Research
- ReSeed: 0.002% extra params in first denoising step restore diversity lost to distillation — kwangmoo_yi · 2026-11-05
- ReSeed: tiny first-step biases restore diversity lost in diffusion distillation — kwangmoo_yi · 2026-11-05
- Anthropic paper: poisoning 0.1% of pretraining data backdoors agents into running attacker code — jiaxinwen22 · 2026-11-05
- Weight space programming spawns a Rube Goldberg machine where parts have tiny brains — _ScottCondron · 2026-11-05
- Claude read all 3,096 ASHG abstracts to surface 33 human genetics highlights — anshulkundaje · 2026-11-05
- Stanford's James Zou on AI agents entering the wet lab, in Nature — james_y_zou · 2026-11-05