2025
Statistics Colloquium: Victor Veitch
11:30 am–12:30 pm Jones 303
Victor Veitch, Department of Statistics and the Data Science Institute, University of Chicago
Title: Statistical Views on LLM post training
Abstract: Typically, large language models are refined with a post training procedure aimed at biasing their outputs to have desirable properties—-helpfulness, harmlessness, factualness, and so forth. These desiderata are often elicited by pairwise comparisons of LLM responses. This comparative reward signal creates some subtleties in how post training should be conducted. I’ll discuss some ways of formalizing the goal of post training and methods for achieving these goals.
Statistics Colloquium: Sara Algeri
11:30 am–12:30 pm Jones 303
Sara Algeri, School of Statistics, University of Minnesota
“When Pearson’s Chi-square and other divisible statistics are not goodness-of-fit tests”
Joint Colloquium with DSI: Jeremy Bernstein
11:30 am–12:30 pm Jones 303
Jeremy Bernstein, Computer Science and Artificial Intelligence Laboratory, MIT
“Metrized deep learning: Fast & scalable optimization via programmatic theory”
Student Seminar: Minji Park
2:00–2:30 pm Jones 304
Thursday, February 20, 2025, at 2:00 PM, in Jones 304, 5747 S. Ellis Avenue
Master’s Thesis l Presentation
Minji Park, Department of Statistics, The University of Chicago
“Evaluating Hybrid Models for Time Series Demand Forecasting: An Empirical Comparison of ES-RNN and Alternative Models”
Student Seminar: Jianan Rui
2:00–2:30 pm Jones 304
Wednesday, February 19, 2025, at 2:00 PM, in Jones 304, 5747 S. Ellis Avenue
Master’s Thesis l Presentation
Jianan Rui, Department of Statistics, The University of Chicago
“Mixed-effects Location Scale Model and Extensions”
Student Seminar: Qianqian Lei
9:30–10:00 am
Wednesday, February 19, 2025, at 9:30 AM, in Jones 111, 5747 S. Ellis Avenue
Master’s Thesis l Presentation
Qianqian Lei, Department of Statistics, The University of Chicago
“Convergence of Stochastic Gradient Descent with Mixture Learning Rates”
Student Seminar: Zhiyuan Chen
9:00–9:30 am
Wednesday, February 19, 2025, at 9:00 AM, in Jones 111, 5747 S. Ellis Avenue
Master’s Thesis l Presentation
Zhiyuan Chen, Department of Statistics, The University of Chicago
“Mixed-effects Location Scale Model and Extensions”
Student Seminar: Jiangchuan Du
2:00–2:30 pm Jones 111
Tuesday, February 18, 2025, at 2:00 PM, in Jones 111, 5747 S. Ellis Avenue
Master’s Thesis l Presentation
Jiangchuan Du, Department of Statistics, The University of Chicago
“Design-based sharp inference for quantiles of individual treatment effects”
Student Seminar: Kyong Ju Lee
2:00–2:30 pm Jones 111
Monday, February 17, 2025, at 2:00 PM, in Jones 111, 5747 S. Ellis Avenue
Master’s Thesis l Presentation
Kyong Ju Lee, Department of Statistics, The University of Chicago
“Doublet Detection via Noisy PU Learning”
Joint Colloquium with DSI: Alex Damian
11:30 am–12:30 pm Jones 303
Joint Colloquium with the Data Science Institute (DSI)
Alex Damian, Program in Applied & Computational Mathematics, Princeton University
“Foundations of Deep Learning: Optimization and Representation Learning”