UniXGrad: A universal, adaptive algorithm with optimal guarantees for constrained optimization

Ali Kavis, Kfir Y. Levy, Francis Bach, Volkan Cevher

Research output: Contribution to journalConference articlepeer-review


We propose a novel adaptive, accelerated algorithm for the stochastic constrained convex optimization setting. Our method, which is inspired by the Mirror-Prox method, simultaneously achieves the optimal rates for smooth/non-smooth problems with either deterministic/stochastic first-order oracles. This is done without any prior knowledge of the smoothness nor the noise properties of the problem. To the best of our knowledge, this is the first adaptive, unified algorithm that achieves the optimal rates in the constrained setting. We demonstrate the practical performance of our framework through extensive numerical experiments.

Original languageEnglish
JournalAdvances in Neural Information Processing Systems
StatePublished - 2019
Event33rd Annual Conference on Neural Information Processing Systems, NeurIPS 2019 - Vancouver, Canada
Duration: 8 Dec 201914 Dec 2019

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing


Dive into the research topics of 'UniXGrad: A universal, adaptive algorithm with optimal guarantees for constrained optimization'. Together they form a unique fingerprint.

Cite this