Ashok Cutkosky

activity

20092026

most citedMomentum-Based Variance Reduction in Non-Convex SGD

30 citations · 102 across the 35 of their papers we have counts for

collaborators

Showing 2019 · stat.MLShow all

3 papers · 2 filters

stat.ML2019★ 1 cited

Anytime Online-to-Batch Conversions, Optimism, and Acceleration

Ashok Cutkosky

A standard way to obtain convergence guarantees in stochastic convex optimization is to run an online learning algorithm and then output the average of its iterates: the actual ite…

stat.ML2019★ 1 cited

Artificial Constraints and Lipschitz Hints for Unconstrained Online Learning

Ashok Cutkosky

We provide algorithms that guarantee regret $R_{T} (u) \leq O ~ (G ∥ u ∥^{3} + G (∥ u ∥ + 1) T)$ or $R_{T} (u) \leq O ~ (G ∥ u ∥^{3} T^{1/3} + G T^{1/3} + G ∥ u ∥ T)$ for online conve…

stat.ML2019★ 3 cited

Combining Online Learning Guarantees

Ashok Cutkosky

We show how to take any two parameter-free online learning algorithms with different regret guarantees and obtain a single algorithm whose regret is the minimum of the two base alg…