Friday, October 29th – Saturday, October 30th 2021 Computer Science Building Room 104, Princeton University The Princeton Day of Statistics (PDS) is a workshop on the various aspects of the frontiers of statistics. The workshop intends to bring top researchers together to define and expand the frontiers of statistics. It also provides a focal venue for senior and junior researchers to discuss and outline emerging problems in their fields and to lay the groundwork for future collaborations. Organizers Matias Cattaneo, Co-Chair Jianqing Fan, Co-Chair Boris Hanin Jason Klusowski Sanjeev Kulkarni Miklos Racz Elizaveta Rebrova Speakers Sivaraman Balakrishnan Joan Bruna Edgar Dobriban Anna Gilbert Jianhua Hu Zheng (Tracy) Ke Samory Kpotufe Ian McKeague Konrad Menzel Jonathan Niles-Weed Alessandro Rinaldo Soledad Villar Andre Wibisono Ming Yuan Nancy Zhang Schedule Friday, October 29th, 2021 Time Speaker or Event Topic or Activity 9:00 am – 9:40 am Zheng (Tracy) Ke, Harvard University Power Analysis and Phase Transitions for FDR Control Methods 9:40 am – 10:20 am Ming Yuan, Columbia University Low Rank Tensor Methods in High Dimensional Data Analysis 10:40 am – 11:20 am Anna Gilbert, Yale University How can classical multidimensional scaling go wrong? 11:20 am – 12:00 pm Soledad Villar, Johns Hopkins University Dimensionality reduction, regularization, and generalization in overparameterized regressions 2:00 pm – 2:40 pm Jianhua Hu, Columbia University New statistical development in biomedical applications 2:40 pm – 3:20 pm Alessandro Rinaldo, Carnegie Mellon University A modern look at inference in linear regression: model-free validity, normal approximations and high dimensions 3:40 pm – 4:20 pm Sivaraman Balakrishnan, Carnegie Mellon University Two methods for assumption-light inference 4:20 pm – 5:00 pm Joan Bruna, New York University Statistical-to-Computational Gaps in learning single periodic neurons Saturday, October 30, 2021 Time Speaker or Event Topic or Activity 9:00 am – 9:40 am Ian McKeague, Columbia University Fallacies of selection: challenges in post-selection inference 9:40 am – 10:20 am Jonathan Niles-Weed, New York University Towards practical estimation of Brenier maps 10:20 am – 11:00 am Nancy Zhang, University of Pennsylvania DNA Copy Number Profiling from Bulk Tissues to Single Cells 11:20 am – 12:00 pm Samory Kpotufe, Columbia University Some Recent Insights on Transfer and Multitask Learning 12:00 pm – 12:40 pm Edgar Dobriban, University of Pennsylvania Comparing Classes of Estimators: When does Gradient Descent Beat Ridge Regression in Linear Models? 2:00 pm – 2:40 pm Konrad Menzel, New York University Central Limit Theory for Models of Strategic Network Formation 2:40 pm – 3:20 pm Andre Wibisono, Yale University On Bias and Discretization: Sampling under Isoperimetry via Langevin Algorithm