Pan Xu

Research Interests

Artificial Intelligence, Data Science, Optimization, Reinforcement Learning, and High Dimensional Statistics

Bio

My research is centered around Machine Learning, with broad interests in the areas of Artificial Intelligence, Data Science, Optimization, Reinforcement Learning, High Dimensional Statistics, and their applications to real-world problems including Bioinformatics and Healthcare. My research goal is to develop computationally- and data-efficient machine learning algorithms with both strong empirical performance and theoretical guarantees.

Education

• Ph.D. University of California, Los Angeles, 2021

Positions

• Assistant Professor of Biostatistics & Bioinformatics
• Assistant Professor of Computer Science
• Assistant Professor in the Department of Electrical and Computer Engineering

Courses Taught

• COMPSCI 391: Independent Study
• BIOSTAT 825: Foundation of Reinforcement Learning

Publications

• Tang C, Liu Z, Xu P. Robust Offline Reinforcement Learning with Linearly Structured $f$-Divergence Regularization. 2024.
• Wang R, Yang Y, Liu Z, Zhou D, Xu P. Return Augmented Decision Transformer for Off-Dynamics Reinforcement Learning. 2024.
• Liu Z, Wang W, Xu P. Upper and Lower Bounds for Distributionally Robust Off-Dynamics Reinforcement Learning. 2024.
• Xu P. Efficient and robust sequential decision making algorithms. AI Magazine. 2024 Sep 1;45(3):376–85.
• Yang Y, Xu P. Pre-trained Language Models Improve the Few-shot Prompt Ability of Decision Transformer. 2024.
• Ren X, Jin T, Xu P. Optimal Batched Linear Bandits. 2024.
• Lopez VK, Cramer EY, Pagano R, Drake JM, O’Dea EB, Adee M, et al. Challenges of COVID-19 Case Forecasting in the US, 2020-2021. PLoS Comput Biol. 2024 May;20(5):e1011200.
• Hsu H-L, Wang W, Pajic M, Xu P. Randomized Exploration in Cooperative Multi-Agent Reinforcement Learning. 2024.
• Jin T, Hsu HL, Chang W, Xu P. Finite-Time Frequentist Regret Bounds of Multi-Agent Thompson Sampling on Sparse Hypergraphs. In: Proceedings of the AAAI Conference on Artificial Intelligence. 2024. p. 12956–64.
• Liu Z, Xu P. Minimax Optimal and Computationally Efficient Algorithms for Distributionally Robust Offline Reinforcement Learning. 2024.
• Liu Z, Xu P. Distributionally Robust Off-Dynamics Reinforcement Learning: Provable Efficiency with Linear Function Approximation. In: Proceedings of Machine Learning Research. 2024. p. 2719–27.
• Qin Z, Liu Z, Xu P. Convergence of Sign-based Random Reshuffling Algorithms for Nonconvex Optimization. 2023.
• Jin T, Yang Y, Tang J, Xiao X, Xu P. Optimal Batched Best Arm Identification. 2023.
• Shen Y, Xu P, Zavlanos MM. Wasserstein Distributionally Robust Policy Evaluation and Learning for Contextual Bandits. 2023.
• Zhang Y, Qu G, Xu P, Lin Y, Chen Z, Wierman A. Global Convergence of Localized Policy Iteration in Networked Multi-Agent Reinforcement Learning. In: SIGMETRICS 2023 - Abstract Proceedings of the 2023 ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems. 2023. p. 83–4.
• Queerinai OO, Ovalle A, Subramonian A, Singh A, Voelcker C, Sutherland DJ, et al. Queer In AI: A Case Study in Community-Led Participatory AI. In: ACM International Conference Proceeding Series. 2023. p. 1882–95.
• Shea K, Borchering RK, Probert WJM, Howerton E, Bogich TL, Li S-L, et al. Multiple models for outbreak decision support in the face of uncertainty. Proc Natl Acad Sci U S A. 2023 May 2;120(18):e2207537120.
• Zhang Y, Qu G, Xu P, Lin Y, Chen Z, Wierman A. Global Convergence of Localized Policy Iteration in Networked Multi-Agent Reinforcement Learning. Proceedings of the ACM on Measurement and Analysis of Computing Systems. 2023 Feb 28;7(1).
• Yang Z, Guo Y, Xu P, Liu A, Anandkumar A. Distributionally Robust Policy Gradient for Offline Contextual Bandits. In: Proceedings of Machine Learning Research. 2023. p. 6443–62.
• Jin T, Yang X, Xiao X, Xu P. Thompson Sampling with Less Exploration is Fast and Optimal. In: Proceedings of Machine Learning Research. 2023. p. 15239–61.
• Cramer EY, Huang Y, Wang Y, Ray EL, Cornell M, Bracher J, et al. The United States COVID-19 Forecast Hub dataset. Sci Data. 2022 Aug 1;9(1):462.
• Cramer EY, Ray EL, Lopez VK, Bracher J, Brennen A, Castro Rivadeneira AJ, et al. Evaluation of individual and ensemble probabilistic forecasts of COVID-19 mortality in the United States. Proc Natl Acad Sci U S A. 2022 Apr 12;119(15):e2113561119.
• Wu Y, Jin T, Lou H, Xu P, Farnoud F, Gu Q. Adaptive Sampling for Heterogeneous Rank Aggregation from Noisy Pairwise Comparisons. In: Proceedings of Machine Learning Research. 2022. p. 11014–36.
• Lou H, Jin T, Wu Y, Xu P, Gu Q, Farnoud F. Active Ranking without Strong Stochastic Transitivity. In: Advances in Neural Information Processing Systems. 2022.
• Xu P, Zheng H, Mazumdar E, Azizzadenesheli K, Anandkumar A. Langevin Monte Carlo for Contextual Bandits. In: Proceedings of Machine Learning Research. 2022. p. 24830–50.
• Jin T, Xu P, Xiao X, Anandkumar A. Finite-Time Regret of Thompson Sampling Algorithms for Exponential Family Multi-Armed Bandits. In: Advances in Neural Information Processing Systems. 2022.
• Jin T, Xu P, Shi J, Xiao X, Gu Q. MOTS: Minimax Optimal Thompson Sampling. In: Proceedings of Machine Learning Research. 2021. p. 5074–83.
• Zou D, Xu P, Gu Q. Faster Convergence of Stochastic Gradient Langevin Dynamics for Non-Log-Concave Sampling. In: Proceedings of Machine Learning Research. 2021. p. 1152–62.
• Jin T, Tang J, Xu P, Huang K, Xiao X, Gu Q. Almost Optimal Anytime Algorithm for Batched Multi-Armed Bandits. In: Proceedings of Machine Learning Research. 2021. p. 5065–73.
• Zhou D, Xu P, Gu Q. Stochastic nested variance reduction for nonconvex optimization. Journal of Machine Learning Research. 2020 May 1;21.
• Zou D, Xu P, Gu Q. Sampling from non-log-concave distributions via stochastic variance-reduced gradient Langevin dynamics. In: AISTATS 2019 - 22nd International Conference on Artificial Intelligence and Statistics. 2020.
• Wu Y, Zhang W, Xu P, Gu Q. A finite-time analysis of two time-scale actor-critic methods. In: Advances in Neural Information Processing Systems. 2020.
• Jin T, Xu P, Gu Q, Farnoud F. Rank aggregation via heterogeneous thurstone preference models. In: AAAI 2020 - 34th AAAI Conference on Artificial Intelligence. 2020. p. 4353–60.
• Xu P, Gu Q. A finite-time analysis of Q-Learning with neural network function approximation. In: 37th International Conference on Machine Learning, ICML 2020. 2020. p. 10486–96.
• Xu P, Gao F, Gu Q. SAMPLE EFFICIENT POLICY GRADIENT METHODS WITH RECURSIVE VARIANCE REDUCTION. In: 8th International Conference on Learning Representations, ICLR 2020. 2020.
• Zhou D, Xu P, Gu Q. Stochastic variance-reduced cubic regularization methods. Journal of Machine Learning Research. 2019 Aug 1;20.
• Xu P, Gao F, Gu Q. An improved convergence analysis of stochastic variance-reduced policy gradient. In: 35th Conference on Uncertainty in Artificial Intelligence, UAI 2019. 2019.
• Zou D, Xu P, Gu Q. Stochastic gradient hamiltonian monte carlo methods with recursive variance reduction. In: Advances in Neural Information Processing Systems. 2019.
• Xu P, Gao F, Gu Q. An Improved Convergence Analysis of Stochastic Variance-Reduced Policy Gradient. In: Proceedings of Machine Learning Research. 2019. p. 541–51.
• Chen J, Xu P, Wang L, Ma J, Gu Q. Covariate adjusted precision matrix estimation via nonconvex optimization. In: 35th International Conference on Machine Learning, ICML 2018. 2018. p. 1464–89.
• Yu Y, Xu P, Gu Q. Third-order smoothness helps: Faster stochastic optimization algorithms for finding local minima. In: Advances in Neural Information Processing Systems. 2018. p. 4525–35.
• Xu P, Wang T, Gu Q. Continuous and discrete-time accelerated stochastic mirror descent for strongly convex functions. In: 35th International Conference on Machine Learning, ICML 2018. 2018. p. 8738–51.
• Zhou D, Xu P, Gu Q. Stochastic variance-reduced cubic regularized Newton method. In: 35th International Conference on Machine Learning, ICML 2018. 2018. p. 9597–606.
• Zhou D, Xu P, Gu Q. Stochastic nested variance reduction for nonconvex optimization. In: Advances in Neural Information Processing Systems. 2018. p. 3921–32.
• Xu P, Wang T, Gu Q. Accelerated stochastic mirror descent: From continuous-time dynamics to discrete-time algorithms. In: International Conference on Artificial Intelligence and Statistics, AISTATS 2018. 2018. p. 1087–96.
• Zou D, Xu P, Gu Q. Subsampled stochastic variance-reduced gradient langevin dynamics. In: 34th Conference on Uncertainty in Artificial Intelligence 2018, UAI 2018. 2018. p. 508–18.
• Zou D, Xu P, Gu Q. Stochastic variance-reduced Hamilton Monte Carlo methods. In: 35th International Conference on Machine Learning, ICML 2018. 2018. p. 9647–56.
• Xu P, Zou D, Chen J, Gu Q. Global convergence of Langevin dynamics based algorithms for nonconvex optimization. In: Advances in Neural Information Processing Systems. 2018. p. 3122–33.
• Xu P, Zhang T, Gu Q. Efficient algorithm for sparse tensor-variate gaussian graphical models via gradient descent. In: Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, AISTATS 2017. 2017.
• Xu P, Ma J, Gu Q. Speeding up latent variable Gaussian graphical model estimation via nonconvex optimization. In: Advances in Neural Information Processing Systems. 2017. p. 1934–45.
• Chaudhry A, Xu P, Gu Q. Uncertainty assessment and false discovery rate control in high-dimensional Granger causal inference. In: 34th International Conference on Machine Learning, ICML 2017. 2017. p. 1120–55.
• Xu P, Gu Q. Semiparametric differential graph models. In: Advances in Neural Information Processing Systems. 2016. p. 1072–80.
• Tian L, Xu P, Gu Q. Forward backward greedy algorithms for multi-task learning with faster rates. In: 32nd Conference on Uncertainty in Artificial Intelligence 2016, UAI 2016. 2016. p. 735–44.
• Jin T, Xu P, Xiao X, Gu Q. Double Explore-then-Commit: Asymptotic Optimality and Beyond. In: Proceedings of Thirty Fourth Conference on Learning Theory. PMLR;
• Xu P, Wen Z, Zhao H, Gu Q. Neural Contextual Bandits with Deep Representation and Shallow Exploration. In.