activity
20152025
most citedUn-regularizing: approximate proximal point and faster stochastic algorithms for empirical risk minimization

67 citations · 176 across the 22 of their papers we have counts for

collaborators
Showing 2020Show all

11 papers · 1 filter

cs.DS2020

Instance Based Approximations to Profile Maximum Likelihood

Nima Anari, Moses Charikar, Kirankumar Shiragur +1

In this paper we provide a new efficient algorithm for approximately computing the profile maximum likelihood (PML) distribution, a prominent quantity in symmetric property estimat…

math.OC2020

Relative Lipschitzness in Extragradient Methods and a Direct Recipe for Acceleration

Michael B. Cohen, Aaron Sidford, Kevin Tian

We show that standard extragradient methods (i.e. mirror prox and dual extrapolation) recover optimal accelerated rates for first-order minimization of smooth convex functions. To…

cs.DS2020

Semi-Streaming Bipartite Matching in Fewer Passes and Optimal Space

Sepehr Assadi, Arun Jambulapati, Yujia Jin +2

We provide -pass semi-streaming algorithms for computing -approximate maximum cardinality matchings in bipartite graphs. Our most efficient methods ar…

math.OC2020

Large-Scale Methods for Distributionally Robust Optimization

Daniel Levy, Yair Carmon, John C. Duchi +1

We propose and analyze algorithms for distributionally robust optimization of convex losses with conditional value at risk (CVaR) and divergence uncertainty sets. We prove th…

cs.DS20201 cited

Coordinate Methods for Matrix Games

Yair Carmon, Yujia Jin, Aaron Sidford +1

We develop primal-dual coordinate methods for solving bilinear saddle-point problems of the form which contain linear p…

cs.LG202020 cited

Efficiently Solving MDPs with Stochastic Mirror Descent

Yujia Jin, Aaron Sidford

We present a unified framework based on primal-dual stochastic mirror descent for approximately solving infinite-horizon Markov decision processes (MDPs) given a generative model.…