Contains slides from my talks.
- On the Role of Nonparametric Statistics in the Age of AI, Noether Award Lecture, Aug 2024, JSM.
- Simultaneous Conformal Prediction of Missing Outcomes with Propensity Score ε-Discretization, Aug 2024, JSM.
- Uncertainty Quantification for Machine Learning, May 2024, U Michigan Curriculum Workshop.
- A Theory of Non-Linear Feature Learning with One Gradient Step in Two-Layer Neural Networks, Lausanne Event on ML Theory (Leman-Th), Lausanne, May 2024.
- Uncertainty Quantification for Machine Learning via PAC Prediction Sets, January 2024, USC SEEDS Conference.
- Efficient and Multiply Robust Risk Estimation under General Forms of Dataset Shift, October 2023, UCSF Biostatistics & Bioinformatics Seminar.
- A Framework for Statistical Inference via Randomized Algorithms, July 2023, ISI WSC 2023, Ottawa.
- Consistency of invariance-based randomization tests, June 2022, ICSA 2022 China Conference.
- T-Cal: An optimal test for the calibration of predictive models, March 2022, Yale, MIT, UC Berkeley, TOPML Conference, Simons Institute NYC.
- Talk for PhD Admits, March 2022, Wharton Stats & Data Science PhD admit day.
- Comparing Classes of Estimators: When does Gradient Descent Beat Ridge Regression in Linear Models?, October 2021, 7th Princeton Day of Statistics.
- The calculus of deterministic equivalents and its applications to high-dimensional statistics, September 2021, ICSA Applied Statistics Symposium, MCP 2021.
- Deep Learning in Statistics: Practical Challenges, May 2021, 2021 Symposium on Data Science & Statistics, Short Course on Deep Learning in Statistics
- What causes the test error? Going beyond bias-variance via ANOVA, May 2021, NSF-Simons Mathematical and Scientific Foundations of Deep Learning Journal Club
- HYPER: Flexible and effective pooled testing via hypergraph factorization, Apr 2021, Princeton IDEAS seminar
- What causes adversarial examples?, Nov 2020, UPenn, ARO MURI reading group
- On the statistical foundations of adversarially robust learning, Oct 2020, Wharton, UPenn, Statistics seminar.
- Asymptotic perspectives on sketching, Wisconsin SILO 2020.
- Discussion on Theoretical Advances in Deep Learning, JSM 2020, Discussant at Session on Theoretical Advances in Deep Learning, organized by Po-Ling Loh.
- The Implicit Regularization of Stochastic Gradient Flow for Least Squares, JSM 2020.
- Ridge Regression: Structure, Cross-Validation, and Sketching, ICLR 2020.
- Understanding Data Augmentation for Deep Learning and Beyond, Denver, JSM 2019.
- A New Theory for Sketching in Linear Regression, Montreal, 2019.
- How to deal with big data? Understanding large-scale distributed regression, Chicago, 2018. Penn 2018.
- Statistics, Data Science, and Machine Learning: A Very Brief Introduction, Penn, 2018. For high school students.
- Deterministic parallel analysis: an improved method for selecting factors and principal components, Paris, France 2017. JSM 2018, Vancouver.
- Optimal prediction in the linearly transformed spiked model, Georgia Tech 2017, Atlanta. JSM 2017, Baltimore
- Weighted multiple testing by convex optimization, Xth MCP 2017, Riverside
- ePCA. Exponential Family PCA, Stanford Statistics Seminar, Feb 2017, Stanford University
- Computation, statistics, and random matrix theory, Harvard Probability and Random Matrix Theory Seminar, Oct 2016, Harvard University
- Optimal detection of principal components in high dimensional data, Stanford Statistics Seminar, Aug 2016, Stanford University. 3rd ISNPS conference, June 2016, Avignon. IDEAS seminar May 2016, Princeton
- Multiple testing with prior information identifies loci for exceptional longevity, poster at Big Data in Biomedicine, May 2016, Stanford
- High-dimensional asymptotics of prediction: ridge regression, ML Reading Group, October 2015, Stanford
- Optimal multiple testing with prior information, IXth MCP 2015, Hyderabad