174 citations · 680 across the 52 of their papers we have counts for
6 papers · 1 filter
Uniform Convergence of Gradients for Non-Convex Learning and Optimization
Dylan J. Foster, Ayush Sekhari, Karthik Sridharan
We investigate 1) the rate at which refined properties of the empirical risk---in particular, gradients---converge to their population counterparts in standard non-convex learning…
Contextual bandits with surrogate losses: Margin bounds and efficient algorithms
Dylan J. Foster, Akshay Krishnamurthy
We use surrogate losses to obtain several new regret bounds and new algorithms for contextual bandit learning. Using the ramp loss, we derive new margin-based regret bounds in term…
Online Learning: Sufficient Statistics and the Burkholder Method
Dylan J. Foster, Alexander Rakhlin, Karthik Sridharan
We uncover a fairly general principle in online learning: If regret can be (approximately) expressed as a function of certain "sufficient statistics" for the data sequence, then th…
Practical Contextual Bandits with Regression Oracles
Dylan J. Foster, Alekh Agarwal, Miroslav Dudík +2
A major challenge in contextual bandits is to design general-purpose algorithms that are both practically useful and theoretically well-founded. We present a new technique that has…
Logistic Regression: The Importance of Being Improper
Dylan J. Foster, Satyen Kale, Haipeng Luo +2
Learning linear predictors with the logistic loss---both in stochastic and online settings---is a fundamental task in machine learning and statistics, with direct connections to cl…
Parameter-free online learning via model selection
Dylan J. Foster, Satyen Kale, Mehryar Mohri +1
We introduce an efficient algorithmic framework for model selection in online learning, also known as parameter-free online learning. Departing from previous work, which has focuse…