preprints in submission/under revision
- Efficient reductions from a Gaussian source and applications to statistical-computational tradeoffs
with Mengqi Lou and Guy Bresler - Predictive inference for time series: why is split conformal effective despite temporal dependence?
with Rina Foygel Barber - State evolution beyond first-order methods I: Rigorous predictions and finite-sample guarantees
with Michael Celentano, Chen Cheng, and Kabir Verchand - Estimating stationary mass, frequency by frequency
with Milind Nakul and Vidya Muthukumar - Modeling and correcting bias in sequential evaluation
with Jingyan Wang
journal publications
- Accurate, provable, and fast polychromatic tomographic reconstruction: A variational inequality approach
with Mengqi Lou, Kabir Verchand, and Sara Fridovich-Keil
SIAM Journal on Imaging Sciences (SIIMS), to appear - Hyperparameter tuning via trajectory predictions: Stochastic prox-linear methods in matrix sensing
with Mengqi Lou and Kabir Verchand
Mathematical Programming Ser. B, Sep 2025 - Computationally efficient reductions between some statistical models
with Mengqi Lou and Guy Bresler
IEEE Transactions on Information Theory, Sep 2025 - A dual accelerated method for online distributed averaging: From consensus to distributed policy evaluation
with Sheng Zhang and Justin Romberg
IEEE Transactions on Automatic Control, April 2025 - Just Wing It: Near-optimal estimation of missing mass in a Markovian sequence
with Vidya Muthukumar and Andrew Thangaraj
Journal of Machine Learning Research, Oct 2024 - Alternating minimization for generalized rank one matrix sensing: Sharp predictions from a random initialization
with Kabir Chandrasekher and Mengqi Lou
Information and Inference: A Journal of the IMA, Sep 2024 - Do algorithms and barriers for sparse principal component analysis extend to other structured settings?
with Guanyi Wang and Mengqi Lou
IEEE Transactions on Signal Processing, Jul 2024 - Optimal and instance-dependent guarantees for Markovian linear stochastic approximation
with Wenlong Mou, Martin J. Wainwright, and Peter L. Bartlett
Mathematical Statistics and Learning, Apr 2024 - Accelerated and instance-optimal policy evaluation with linear function approximation
with Tianjiao Li and Guanghui Lan
SIAM Journal on Mathematics of Data Science (SIMODS), Mar 2023 - Sharp global convergence guarantees for iterative nonconvex optimization with random data
with Kabir Chandrasekher and Christos Thrampoulidis
Annals of Statistics, Feb 2023
Runner-up, Best Paper Prize for Young Researchers in Continuous Optimization, Mathematical Optimization Society - Optimal oracle inequalities for solving projected fixed-point equations, with applications to policy evaluation
with Wenlong Mou and Martin J. Wainwright
Mathematics of Operations Research, Dec 2022 - Max-affine regression: Parameter estimation for Gaussian designs
with Avishek Ghosh, Adityanand Guntuboyina, and Kannan Ramchandran
IEEE Transactions on Information Theory, Mar 2022 - Isotonic regression with unknown permutations: Statistics, computation, and adaptation
with Richard J. Samworth
Annals of Statistics, Feb 2022 - Is temporal difference learning optimal? An instance-dependent analysis
with Koulik Khamaru, Feng Ruan, Martin J. Wainwright, and Michael I. Jordan
SIAM Journal on Mathematics of Data Science (SIMODS), Oct 2021
ICML 2020 workshop on theoretical foundations of reinforcement learning - Single-index models in the high signal regime
with Dean P. Foster
IEEE Transactions on Information Theory, June 2021 - Instance-dependent $\ell_\infty$-bounds for policy evaluation in tabular reinforcement learning
with Martin J. Wainwright
IEEE Transactions on Information Theory, Jan 2021 - Towards Optimal Estimation of Bivariate Isotonic Matrices with Unknown Permutations
with Cheng Mao and Martin J. Wainwright
Annals of Statistics, Dec 2020 - Worst-case vs average-case design for estimation from partial pairwise comparisons
with Cheng Mao, Vidya Muthukumar, Martin J. Wainwright, and Thomas A. Courtade
Annals of Statistics, April 2020 - Derivative-Free Methods for Policy Optimization: Guarantees for Linear Quadratic Systems
with Dhruv Malik, Kush Bhatia, Koulik Khamaru, Peter L. Bartlett, and Martin J. Wainwright
Journal of Machine Learning Research, Jan 2020
NIPS 2017 workshop on learning on distributions, functions, graphs and groups (oral) - Existence of Stein kernels under spectral gap, and discrepancy bounds
with Thomas A. Courtade and Max Fathi
Annales de l’Institut Henri Poincare, May 2019 - Quantitative stability of the Entropy Power Inequality
with Thomas A. Courtade and Max Fathi
IEEE Transactions on Information Theory, Aug 2018 - Linear regression with shuffled data: Statistical and computational limits of permutation recovery
with Martin J. Wainwright and Thomas A. Courtade
IEEE Transactions on Information Theory, May 2018 - The effect of local decodability constraints on variable-length compression
with Thomas A. Courtade
IEEE Transactions on Information Theory, April 2018 - Optimally approximating the coverage lifetime of wireless sensor networks
with Vivek Kumar Bagaria and Rahul Vaze
IEEE/ACM Transactions on Networking, Feb 2017 - On the complexity of making a distinguished vertex minimum or maximum degree by vertex deletion
with Sounaka Mishra and N. Safina Devi
Journal of Discrete Algorithms, July 2015
conference publications
- Estimating stationary mass, frequency by frequency
with Milind Nakul and Vidya Muthukumar
COLT 2025 (extended abstract) - Accurate, provable, and fast nonlinear tomographic reconstruction: A variational inequality approach
with Mengqi Lou, Kabir Verchand, and Sara Fridovich-Keil
Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine (Fully3D) 2025, superseded by journal version - Computationally efficient reductions between some statistical models
with Mengqi Lou and Guy Bresler
ALT 2025 (extended abstract), superseded by journal version
Outstanding paper award - Learning the eye of the beholder: Statistical modeling and estimation for personalized color perception
with Xuanzhou (Lorraine) Chen, Austin Xu, and Jingyan Wang
Allerton 2024 - One shot inverse reinforcement learning for stochastic linear bandits
with Etash Guha, Jim James, Krishna Acharya, and Vidya Muthukumar
UAI 2024 - Alternating minimization for generalized rank one matrix sensing: Sharp predictions from a random initialization
with Kabir Chandrasekher and Mengqi Lou
ALT 2024 (extended abstract), superseded by journal version - Perceptual adjustment queries and an inverted measurement paradigm for low-rank metric learning
with Austin Xu, Andrew McRae, Jingyan Wang, and Mark A. Davenport
NeurIPS 2023 - Sharp analysis of EM for learning mixtures of pairwise differences
with Abhishek Dhawan and Cheng Mao
COLT 2023 - Modeling and correcting bias in sequential evaluation
with Jingyan Wang
EC 2023 (extended abstract) - A dual accelerated method for a class of distributed optimization problems
with Sheng Zhang and Justin Romberg
CDC 2022, superseded by journal version - Optimal and instance-dependent guarantees for Markovian linear stochastic approximation
with Wenlong Mou, Martin J. Wainwright, and Peter L. Bartlett
COLT 2022 (extended abstract), superseded by journal version - Learning from an exploring demonstrator: Optimal reward estimation for bandits
with Wenshuo Guo, Kumar Krishna Agrawal, Aditya Grover, and Vidya Muthukumar
AISTATS 2022
ICML 2021 workshop on Human-AI Collaboration in Sequential Decision-Making (spotlight)
- Preference learning along multiple criteria: A game-theoretic perspective
with Kush Bhatia, Peter L. Bartlett, Anca D. Dragan, and Martin J. Wainwright
NeurIPS 2020
ICML 2020 workshop on theoretical foundations of reinforcement learning (oral) - Max-affine regression with universal parameter estimation for small-ball designs
with Avishek Ghosh, Adityanand Guntuboyina, and Kannan Ramchandran
ISIT 2020 - A family of Bayesian Cramer-Rao bounds, and consequences for log-concave priors
with Efe Aras, Kuan-Yun Lee, and Thomas A. Courtade
ISIT 2019 - Derivative-free methods for policy optimization: Guarantees for linear quadratic systems
with Dhruv Malik, Kush Bhatia, Koulik Khamaru, Peter L. Bartlett, and Martin J. Wainwright
AISTATS 2019, superseded by journal version - Breaking the $1/\sqrt{n}$ barrier: Faster rates for permutation-based models in polynomial time
with Cheng Mao and Martin J. Wainwright
COLT 2019 (extended abstract), superseded by journal version - Gradient diversity: A key ingredient for scalable distributed learning
with Dong Yin, Max Lam, Dimitris Papailiopoulos, Kannan Ramchandran, and Peter L. Bartlett
AISTATS 2018
NIPS 2017 OPT-ML workshop (oral) - Wasserstein stability of the entropy power inequality for log-concave densities
with Thomas A. Courtade and Max Fathi
ISIT 2017, superseded by journal version - Denoising linear models with permuted data
with Martin J. Wainwright and Thomas A. Courtade
ISIT 2017 - Linear regression with an unknown permutation: Statistical and computational limits
with Martin J. Wainwright and Thomas A. Courtade
Allerton 2016, superseded by journal version - Compressing sparse sequences under local decodability constraints
with Thomas A. Courtade
ISIT 2015, superseded by journal version - The online disjoint set cover problem and its applications
with Vivek Kumar Bagaria and Rahul Vaze
INFOCOM 2015 - Maximizing utility among selfish users in social groups
with Vivek Kumar Bagaria and Rahul Vaze
NCC 2014
Best Paper Award, Networks Track