4 papers
A Unified Theory of H-Duality in First-Order Methods
Kevin Shu, Alex L. Wang
We provide two complementary explanations of H-duality in smooth strongly convex optimization and contractive fixed-point problems. H-duality refers to the phenomenon where the wor…
Beyond Minimax Optimality: A Subgame Perfect Gradient Method
Benjamin Grimmer, Kevin Shu, Alex L. Wang
The study of convex optimization has historically been concerned with worst-case convergence rates. The development of the Optimized Gradient Method (OGM), due to \citet{drori2012P…
Composing Optimized Stepsize Schedules for Gradient Descent
Benjamin Grimmer, Kevin Shu, Alex L. Wang
Recent works by Altschuler and Parrilo and the authors have shown that it is possible to accelerate the convergence of gradient descent on smooth convex functions, even without mom…
A Strengthened Conjecture on the Minimax Optimal Constant Stepsize for Gradient Descent
Benjamin Grimmer, Kevin Shu, Alex L. Wang
Drori and Teboulle [4] conjectured that the minimax optimal constant stepsize for N steps of gradient descent is given by the stepsize that balances performance on Huber and quadra…