PUBLICATIONS

Unless specified otherwise, authors are listed in alphabetical order. The * sign indicates equal contribution.

Preprints

2022

Causal Imitation Learning under Temporally Correlated Noise

Gokul Swamy, Sanjiban Choudhury, J. Drew Bagnell, and Z. S. W. (contributional order)
The 39th International Conference on Machine Learning (ICML'22 Long presentation)
arXiv


Strategic Instrumental Variable Regression: Recovering Causal Relationships From Strategic Responses

Keegan Harris, Daniel Ngo*, Logan Stapleton*, Hoda Heidari, Z. S. W. (contributional order)
The 39th International Conference on Machine Learning (ICML'22)
arXiv


Personalization Improves Privacy-Accuracy Tradeoffs in Federated Optimization

Alberto Bietti, Chen-Yu Wei, Miro Dudik, John Langford, Z. S. W. (contributional order)
The 39th International Conference on Machine Learning (ICML'22)
arXiv


Information Discrepancy in Strategic Learning

with Yahav Bechavod, Chara Podimata, and Juba Ziani
The 39th International Conference on Machine Learning (ICML'22)
arXiv


Understanding Clipping for Federated Learning: Convergence and Client-Level Differential Privacy

Xinwei Zhang, Xiangyi Chen, Mingyi Hong, Z. S. W., and Jinfeng Yi (contributional order)
The 39th International Conference on Machine Learning (ICML'22)
arXiv


Improved Regret for Differentially Private Exploration in Linear MDP

with Daniel Ngo and Giuseppe Vietri
The 39th International Conference on Machine Learning (ICML'22)
arXiv


Constrained Variational Policy Optimization for Safe Reinforcement Learning

Zuxin Liu, Zhepeng Cen, Vladislav Isenbaev, Wei Liu, Z. S. W., Bo Li, and Ding Zhao (contributional order)
The 39th International Conference on Machine Learning (ICML'22)
arXiv


How Child Welfare Workers Reduce Racial Disparities in Algorithmic Decisions

Hao-Fei Cheng*, Logan Stapleton*, Anna Kawakami, Venkat Sivaraman, Yanghuidi Cheng, Diana Qing, Adam Perer, Ken Holstein, Z. S. W., and Haiyi Zhu (contributional order)
The 2022 ACM CHI Conference on Human Factors in Computing Systems (CHI'22)
ACM Extended Analysis Short Talk Blog


Improving Human-AI Partnerships in Child Welfare: Understanding Worker Practices, Challenges, and Desires for Algorithmic Decision Support

Anna Kawakami, Venkat Sivaraman, Hao-Fei Cheng, Logan Stapleton, Yanghuidi Cheng, Diana Qing, Adam Perer, Z. S. W., Haiyi Zhu, and Ken Holstein (contributional order)
The 2022 ACM CHI Conference on Human Factors in Computing Systems (CHI'22)
CHI Best Paper Honorable Mention
ACM arixiv Short Talk


'Why Do I Care What's Similar?' Probing Challenges in AI-Assisted Child Welfare Decision-Making through Worker-AI Interface Design Concepts

Anna Kawakami*, Venkat Sivaraman*, Hao-Fei Cheng, Logan Stapleton, Adam Perer, Z. S. W., Haiyi Zhu, and Ken Holstein (contributional order)
The 2022 ACM conference on Designing Interactive Systems (DIS'22)


Exploring How Machine Learning Practitioners (Try To) Use Fairness Toolkits

Wesley Hanwen Deng, Manish Nagireddy, Michelle Seng Ah Lee, Jatinder Singh, Z. S. W., Ken Holstein, and Haiyi Zhu (contributional order)
The Fifth ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT'22)


Imagining new futures beyond predictive systems in child welfare: A qualitative study with impacted stakeholders

Logan Stapleton, Min Hun Lee, Diana Qing, Marya Wright, Alexandra Chouldechova, Z. S. W., Ken Holstein, and Haiyi Zhu (contributional order)
The Fifth ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT'22)


Bandit Data-Driven Optimization for Crowdsourcing Food Rescue Platforms

Zheyuan Ryan Shi, Z. S. W., Rayid Ghani, and Fei Fang (contributional order)
The 36th AAAI Conference on Artificial Intelligence (AAAI'22)
arXiv AAAI


2021

Iterative Methods for Private Synthetic Data: Unifying Framework and New Methods

with Terrance Liu, and Giuseppe Vietri
The Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS'21)
arXiv


Stateful Strategic Regression

with Keegan Harris, and Hoda Heidari
The Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS'21)
arXiv


Of Moments and Matching: Trade-offs and Treatments in Imitation Learning

Gokul Swamy, Sanjiban Choudhury, J. Drew Bagnell, and Z. S. W. (contributional order)
The 38th International Conference on Machine Learning (ICML'21)
arXiv


Leveraging Public Data for Practical Private Query Release

Terrance Liu, Giuseppe Vietri, Thomas Steinke, Jonathan Ullman, and Z. S. W. (contributional order)
The 38th International Conference on Machine Learning (ICML'21)
Best paper awards from ICLR workshops: Distributed and Private Machine Learning and Synthetic Data Generation.
arXiv


Towards the Unification and Robustness of Perturbation and Gradient Based Explanations

Sushant Agarwal, Shahin Jabbari, Chirag Agarwal*, Sohini Upadhyay*, Z. S. W., and Hima Lakkaraju (contributional order)
The 38th International Conference on Machine Learning (ICML'21)
arXiv


Incentivizing Compliance with Algorithmic Instruments

Daniel Ngo*, Logan Stapleton*, Vasilis Syrgkanis, Z. S. W. (contributional order)
The 38th International Conference on Machine Learning (ICML'21)
arXiv


An Algorithmic Framework for Fairness Elicitation

with Chris Jung, Michael Kearns, Seth Neel, Aaron Roth, and Logan Stapleton
The second annual Symposium on Foundations of Responsible Computing (FORC'21)
arXiv


Private Post-GAN Boosting

Marcel Neunhoeffer, Z. S. W., and Cynthia Dwork (contributional order)
The Ninth International Conference on Learning Representations (ICLR'21)
arXiv


Bypassing the Ambient Dimension: Private SGD with Gradient Subspace Identification

Yingxue Zhou, Z. S. W., and Arindam Banerjee (contributional order)
The Ninth International Conference on Learning Representations (ICLR'21)
arXiv


Learn to Expect the Unexpected: Probably Approximately Correct Domain Generalization

with Vikas K. Garg, Katrina Ligett, and Adam Kalai
The 24th International Conference on Artificial Intelligence and Statistics (AISTATS'21)
arXiv


Gaming Helps! Learning from Strategic Interactions in Natural Dynamics

with Yahav Bechavod, Katrina Ligett, and Juba Ziani
The 24th International Conference on Artificial Intelligence and Statistics (AISTATS'21)
arXiv


Soliciting Stakeholders' Fairness Notions in Child Maltreatment Predictive Systems

Hao-Fei Cheng, Logan Stapleton, Ruiqi Wang, Paige Bullock, Alexandra Chouldechova, Z. S. W., and Haiyi Zhu (contributional order)
The 2021 ACM CHI Conference on Human Factors in Computing Systems (CHI'21)
arXiv


Value Cards: An Educational Toolkits for Teaching Social Impacts of Machine Learning through Deliberation

Hong Shen, Wesley Deng, Aditi Chattopadhyay, Z. S. W., Xu Wang, and Haiyi Zhu (contributional order)
The Fourth ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT'21)
arXiv


2020

Metric-Free Individual Fairness in Online Learning

with Yahav Bechavod and Chris Jung
The Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS'20 Oral)
arXiv


Understanding Gradient Clipping in Private SGD: A Geometric Perspective

Xiangyi Chen, Z. S. W., and Mingyi Hong (contributional order)
The Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS'20 Spotlight)
arXiv


Distributed Training with Heterogeneous Data: Bridging Median- and Mean-Based Algorithms

Xiangyi Chen*, Tiancong Chen*, Haoran Sun, Z. S. W., and Mingyi Hong (contributional order)
The Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS'20)
arXiv


Private Reinforcement Learning with PAC and Regret Guarantees

Giuseppe Vietri, Borja Balle, Akshay Krishnamurthy, and Z. S. W. (contributional order)
The 37th International Conference on Machine Learning (ICML'20)
arXiv


New Oracle-Efficient Algorithms for Private Synthetic Data Release

Giuseppe Vietri, Grace Tian, Mark Bun, Thomas Steinke, and Z. S. W. (contributional order)
The 37th International Conference on Machine Learning (ICML'20)
arXiv


Structured Linear Contextual Bandits: A Sharp and Geometric Smoothed Analysis

Vidyashankar Sivakumar, Z. S. W., and Arindam Banerjee (contributional order)
The 37th International Conference on Machine Learning (ICML'20)
arXiv


Private Query Release Assisted by Public Data

with Raef Bassily, Albert Cheu, Shay Moran, Aleksandar Nikolov, and Jonathan Ullman
The 37th International Conference on Machine Learning (ICML'20)
arXiv


Oracle Efficient Private Non-Convex Optimization

with Seth Neel, Aaron Roth, and Giuseppe Vietri
The 37th International Conference on Machine Learning (ICML'20)
arXiv


Privately Learning Markov Random Fields

Huanyu Zhang, Gautam Kamath*, Janardhan Kulkarni*, and Z. S. W.* (contributional order)
The 37th International Conference on Machine Learning (ICML'20)
arXiv


Locally Private Hypothesis Selection

with Sivakanth Gopi, Gautam Kamath, Janardhan Kulkarni, Aleksandar Nikolov, and Huanyu Zhang
The 33rd Annual Conference on Learning Theory (COLT'20)
arXiv


Keeping Designers in the Loop: Communicating Inherent Algorithmic Trade-offs Across Multiple Objectives

Bowen Yu, Irene Yuan, Loren Terveen, Z. S. W., Jodi Forlizzi, and Haiyi Zhu (contributional order)
The 2020 ACM conference on Designing Interactive Systems (DIS'20)


arXiv


Incentivizing Exploration with Selective Data Disclosure

with Nicole Immorlica, Jieming Mao, and Alex Slivkins
The 21st ACM Conference on Economics and Computation (EC'20)
arXiv


2019

Private Hypothesis Selection

with Mark Bun, Gautam Kamath, and Thomas Steinke
The Thirty-Third Conference on Neural Information Processing Systems (NeurIPS'19)
arXiv


Equal Opportunity in Online Classification with Partial Feedback

with Yahav Bechavod, Katrina Ligett, Aaron Roth, and Bo Waggoner
The Thirty-Third Conference on Neural Information Processing Systems (NeurIPS'19)
arXiv


Locally Private Gaussian Estimation

with Matthew Joseph, Janardhan Kulkarni, and Jieming Mao
The Thirty-Third Conference on Neural Information Processing Systems (NeurIPS'19)
arXiv


Random Qudratic Forms with Dependence: Applications to Restricted Isometry and Beyond

with Arindam Banerjee, Qilong Gu, and Vidyashankar Sivakumar
The Thirty-Third Conference on Neural Information Processing Systems (NeurIPS'19)
arxiv


Privacy-Preserving Generative Deep Neural Networks Support Clinical Data Sharing

Brett Beaulieu-Jones, Z. S. W., Chris Williams, Ran Lee, Sanjeev Bhavnani, James Byrd, and Casey Greene (contributional order)
Circulation: Cardiovascular Quality and Outcomes 2019; 12
bioRxiv Circ link GitHub


How to Use Heuristics for Differential Privacy

with Seth Neel and Aaron Roth
The 60th Annual IEEE Symposium on Foundations of Computer Science (FOCS'19)
arXiv


Fair Regression: Quantitative Definitions and Reduction-Based Algorithms

with Alekh Agarwal and Miro Dudik
The 36th International Conference on Machine Learning (ICML'19)
arXiv GitHub


Locally Private Bayesian Inference for Count Models

Aaron Schein, Z. S. W., Xanda Schofield, Mingyuan Zhou, and Hanna Wallach (contributional order)
The 36th International Conference on Machine Learning (ICML'19)
arXiv


Orthogonal Random Forest for Causal Inference

with Miruna Oprescu and Vasilis Syrgkanis
The 36th International Conference on Machine Learning (ICML'19)
arXiv GitHub


The Perils of Exploration under Competition: A Computational Modeling Approach

with Guy Aridor, Kevin Liu, and Alex Slivkins
The 20th ACM conference on Economics and Computation (EC'19)
arXiv Talk


Bayesian Exploration with Heterogeneous Agents

with Nicole Immorlica, Jieming Mao, and Alex Slivkins
The Web Conference 2019 (TheWebConf'19 w/ Oral presentation)
arXiv MSFT Blog Post


An Empirical Study of Rich Subgroup Fairness for Machine Learning

with Michael Kearns, Seth Neel, and Aaron Roth
The Second Annual ACM Conference on Fairness, Accountability, and Transparency (FAT*'19)
arXiv GitHub


2018

A Smoothed Analysis of the Greedy Algorithm for the Linear Contextual Bandit Problem

with Sampath Kannan, Jamie Morgenstern, Aaron Roth, and Bo Waggoner
The Thirty-Second Conference on Neural Information Processing Systems (NeurIPS'18 Spotlight)
arXiv


Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness

with Michael Kearns, Seth Neel, and Aaron Roth
The 35th International Conference on Machine Learning (ICML'18)
arXiv Penn News Coverage Aaron's Blog Post TCS + talk by MK


Semiparametric Contextual Bandits

Akshay Krishnamurthy, Z. S. W., and Vasilis Syrgkanis (contributional order)
The 35th International Conference on Machine Learning (ICML'18)
arXiv


The Externalities of Exploration and How Data Diversity Helps Exploitation

with Manish Raghavan, Alex Slivkins, and Jenn Wortman Vaughan
The 31st Annual Conference on Learning Theory (COLT'18)
arXiv


Strategic Classification from Revealed Preferences

with Jinshuo Dong, Aaron Roth, Zachary Schutzman, and Bo Waggoner
The 19th ACM conference on Economics and Computation (EC'18)
arXiv


Competing Bandits: Learning under Competition

with Yishay Mansour and Alex Slivkins
The 9th Innovations in Theoretical Computer Science (ITCS'18)
arXiv Talk


2017

Accuracy First: Selecting a Differential Privacy Level for Accuracy-Constrained ERM

with Katrina Ligett, Seth Neel, Aaron Roth, and Bo Waggoner
The Thirty-First Annual Conference on Neural Information Processing Systems (NIPS'17)
arXiv GitHub


Meritocratic Fairness in Cross-Population Selection

with Michael Kearns, and Aaron Roth
The 34th International Conference on Machine Learning (ICML'17)
Proceeding Full Version


Predicting with Distributions

with Michael Kearns
The 30th Annual Conference on Learning Theory (COLT'17)
arXiv


Multidimensional Dynamic Pricing for Welfare Maximization

with Aaron Roth, Alex Slivkins, and Jonathan Ullman
The 18th ACM Conference on Economics and Computation (EC'17)
Invited to a special issue of Transactions on Economics and Computation for EC'17

arXiv Talk


Fairness Incentives for Myopic Agents

with Sampath Kannan, Michael Kearns, Jamie Morgenstern, Mallesh Pai, Aaron Roth, and Rakesh Vohra
The 18th ACM Conference on Economics and Computation (EC'17)
arXiv


2016

Learning from Rational Behavior: Predicting Solutions to Unknown Linear Programs

with Shahin Jabbari, Ryan Rogers, and Aaron Roth
The Thirtieth Annual Conference on Neural Information Processing Systems (NIPS'16)
arXiv


Bayesian Exploration: Incentivizing Exploration in Bayesian Games

with Yishay Mansour, Alex Slivkins, and Vasilis Syrgkanis
The 17th ACM Conference on Economics and Computation (EC'16)
Operations Research 2022
Invited to a special issue of Transactions on Economics and Computation for EC'16

arXiv OR Journal


Private Algorithms for the Protected in Social Network Search

with Michael Kearns, Aaron Roth, and Grigory Yaroslavtsev
In Proceedings of the National Academy of Sciences (PNAS), 2016
PNAS Link arXiv

Press coverage by Quartz Pacific Standard Naked Scientist


Watch and Learn: Optimizing from Revealed Preferences Feedback

with Aaron Roth and Jonathan Ullman
The 48th ACM Symposium on Theory of Computing (STOC'16)
arXiv Sigecom Exchanges


Adaptive Learning with Robust Generalization Guarantees

with Rachel Cummings, Katrina Ligett, Kobbi Nissim, and Aaron Roth
The 29th Annual Conference on Learning Theory (COLT'16)
arXiv


Logarithmic Query Complexity for Approximate Nash Computation in Large Games

with Paul Goldberg and Francisco J. Marmolejo Cossio
The 9th International Symposium on Algorithmic Game Theory (SAGT'16)
In the special issue of Theory of Computing for SAGT'16

Springer Link arXiv


Coordination Complexity: Small Information Coordinating Large Populations

with Rachel Cummings, Katrina Ligett, Jaikumar Radhakrishnan, and Aaron Roth
The 7th Innovations in Theoretical Computer Science (ITCS'16)
arXiv


Jointly Private Convex Programming

with Justin Hsu, Zhiyi Huang, and Aaron Roth
The 27th ACM-SIAM Symposium on Discrete Algorithms (SODA'16)
arXiv


2015

Privacy and Truthful Equilibrium Selection for Aggregative Games

with Rachel Cummings, Michael Kearns, and Aaron Roth
The 11th Conference on Web and Internet Economics (WINE'15)
arXiv


Inducing Approximately Optimal Flow Using Truthful Mediators

with Ryan Rogers, Aaron Roth, and Jonathan Ullman
The 16th ACM Conference on Economics and Computation (EC'15)
arXiv


Accuracy for Sale: Aggregating Data with a Variance Constraint

with Rachel Cummings, Katrina Ligett, Aaron Roth, and Juba Ziani
The 6th Innovations in Theoretical Computer Science (ITCS'15)
Proceeding


Approximately Stable, School Optimal, and Student-Truthful Many-to-One Matchings (via Differential Privacy)

with Sampath Kannan, Jamie Morgenstern, and Aaron Roth
The 26th ACM-SIAM Symposium on Discrete Algorithms (SODA'15)
arXiv Blog Post by Lipton and Regan


2014

Dual Query: Practical Private Query Release for High Dimensional Data

with Marco Gaboardi, Emilio Jesús Gallego Arias, Justin Hsu, and Aaron Roth
The 31st International Conference on Machine Learning (ICML'14)
Journal of Privacy and Confidentiality, 2016

arXiv Talk GitHub


Private Matchings and Allocations

with Justin Hsu, Zhiyi Huang, Aaron Roth, and Tim Roughgarden
The 46th ACM Symposium on Theory of Computing (STOC'14)
SIAM Journal on Computing, 45(6):1953–1984, 2016.

arXiv Talk