Venkata Krishna Pillutla — accepted AI-conference papers

2018

A Smoother Way to Train Structured Prediction Models

Venkata Krishna Pillutla, Vincent Roulet, Sham M. Kakade, Zaid Harchaoui

NeurIPS 2018poster

We present a framework to train a structured prediction model by performing smoothing on the inference algorithm it builds upon. Smoothing overcomes the non-smoothness inherent to the maximum margin structured prediction objective, and paves the way for the use of fast primal gradient-based optimiza…

Cited by 25 Source ↗PDF ↗code ↗Scholar ↗

2017

Data Driven Resource Allocation for Distributed Learning

Travis Dick, Mu Li, Venkata Krishna Pillutla, Colin White, Nina Balcan, Alex Smola

AISTATS 2017poster

In distributed machine learning, data is dispatched to multiple machines for processing. Motivated by the fact that similar data points often belong to the same or similar classes, and more generally, classification rules of high accuracy tend to be “locally simple but globally complex” (Vapnik and…

Cited by 17 Source ↗PDF ↗Scholar ↗