Daniel Russo – Research

Approximation Benefits of Policy Gradient Methods with Aggregated States.
Daniel Russo
Working paper

Global Optimality Guarantees For Policy Gradient Methods
Jalaj Bhandari and Daniel Russo
Working paper
A related note on the linear convergence of policy gradient methods
Talk link

On the Futility of Dynamics in Robust Mechanism Design
Santiago Balseiro, Anthony Kim and Daniel Russo
Electronic Companion
Under review – second round

Worst-Case Regret Bounds For Exploration Via Randomized Value Functions
Daniel Russo
NeurIPS 2019

A Note on the Equivalence of Upper Confidence Bounds and Gittins Indices for Patient Agents
Daniel Russo
Operations Research (to appear)

A Finite Time Analysis of Temporal Difference Learning With Linear Function Approximation
Jalaj Bhandari, Daniel Russo, and Raghav Singal
Operations Research (to appear)
Preliminary version appeared at COLT 2018
Short talk link

A Tutorial on Thompson Sampling
Daniel Russo, Benjamin Van Roy, Abbas Kazerouni, Ian Osband, and Zheng Wen
Foundations and Trends in Machine Learning, Vol. 11, No. 1, pp. 1-96, 2018. (code)

Satisficing in Time-Sensitive Bandit Learning
Daniel Russo and Benjamin Van Roy
Under review – second round

Deep Exploration via Randomized Value Functions
Ian Osband, Daniel Russo, Zheng Wen, and Benjamin Van Roy
Journal of Machine Learning Research, 2019

Improving the Expected Improvement Algorithm
Chao Qin, Diego Klabjan and Daniel Russo
NeurIPS 2017

Simple Bayesian Algorithms for Best Arm Identification
Daniel Russo
Operations Research, 2020
Prelimnary version appeared in COLT 2016
First place, INFORMS JFIG paper competition.

Controling Bias in Adaptive Data Analysis Using Information Theory
Daniel Russo and James Zou
IEEE Transaction on Information Theory, 2020
Preliminary version appeared at AISTATS 2016 (full oral presentation; top 7% of submissions).

Learning to Optimize Via Information Directed Sampling
Daniel Russo and Benjamin Van Roy
Operations Research, 2018
Prelimnary version appeared at NeurIPS 2014
First place, INFORMS George Nicholson 2014 student paper competition.

An Information-Theoretic Analysis of Thompson Sampling
Daniel Russo and Benjamin Van Roy
Journal of Machine Learning Research, 2016

Learning to Optimize Via Posterior Sampling
Daniel Russo and Benjamin Van Roy
Mathematics of Operations Research. Vol. 39. No. 4, pp. 1221-1243, 2014.

Eluder Dimension and the Sample Complexity of Optimistic Exploration
Daniel Russo and Benjamin Van Roy
NeurIPS 2013 (full oral presentation; top 1.4% of submissions).

(More) Efficient Reinforcement Learning via Posterior Sampling
Ian Osband, Daniel Russo, and Benjamin Van Roy
NeurIPS 2013.

Welfare-Improving Cascades and the Effect of Noisy Reviews
Nick Arnosti and Daniel Russo
WINE 2013.