# All events

## Wednesday, August 22, 2018

### Uniform rates of the Glivenko-Cantelli convergence and their use in approximating Bayesian inferences

Seminar: Probability Seminar | August 22 | 3-4 p.m. | 1011 Evans Hall

Eugenio Regazzini, Universita degli Studi di Pavia, Italy

This talk deals with suitable quantifications in approximating a probability measure by an “empirical” random probability measure \hat p_n, depending on the first n terms of a sequence \{\xi_i\}_{i\ge1}

of random elements. In the first part, we study the range of oscillation near zero of the p-Wasserstein distance d(p) ....

Based on joint work with Emanuele Dolera. More >

### Rerandomization and Regression Adjustment

Seminar: Neyman Seminar | August 22 | 4-5 p.m. | 1011 Evans Hall

Peng Ding, UC Berkeley

Randomization is a basis for the statistical inference of treatment effects without assumptions on the outcome generating process. Appropriately using covariates further yields more precise estimators in randomized experiments. In his seminal work Design of Experiments, R. A. Fisher suggested blocking on discrete covariates in the design stage and conducting the analysis of covariance (ANCOVA) in... More >

## Tuesday, August 28, 2018

### Seminar 217, Risk Management: Is motor insurance ratemaking going to change with telematics and semi-autonomous vehicles?

Seminar: Risk Seminar | August 28 | 11 a.m.-12:30 p.m. | 1011 Evans Hall

Speaker: Montserrat Guillen, University of Barcelona

Consortium for Data Analytics in Risk

Many automobile insurance companies offer the possibility to monitor driving habits and distance driven by means of telematics devices installed in the vehicles. This provides a novel source of data that can be analysed to calculate personalised tariffs. For instance, drivers who accumulate a lot of miles should be charged more for their insurance coverage than those who make little use of their... More >

## Wednesday, August 29, 2018

### Spectrum of random non-selfadjoint operators

Seminar: Probability Seminar | August 29 | 3-4 p.m. | 1011 Evans Hall

Martin Vogel, UC Berkeley

The spectrum of non-selfadjoint operators can be highly unstable even under very small perturbations. This phenomenon is referred to as "pseudospectral effect".

Traditionally this pseudosepctral effect was considered a drawback since it can be the source of immense numerical errors, as shown for instance in the works of L. N. Trefethen. However, this pseudospectral effect can also be the source... More >

### Likelihood Ratio Test for Stochastic Block Models with Bounded Degrees

Seminar: Neyman Seminar | August 29 | 4-5 p.m. | 1011 Evans Hall

Yang Feng, Columbia University

A fundamental problem in network data analysis is to test whether a network contains statistical significant communities. We study this problem in the stochastic block model context by testing H0: Erdos-Renyi model vs. H1: stochastic block model. This problem serves as the foundation for many other problems including the testing-based methods for determining the number of communities and... More >

## Thursday, August 30, 2018

### Stochastic Gradient Descent: Strong convergence guarantees -- without parameter tuning

Seminar: Neyman Seminar: Special Seminar | August 30 | 4-5 p.m. | 60 Evans Hall

Rachel Ward, UT Austin

Department of Statistics, Department of Mathematics

Stochastic Gradient Descent is the basic optimization algorithm behind powerful deep learning architectures which are becoming increasingly omnipresent in society. However, existing theoretical guarantees of convergence rely on knowing certain properties of the optimization problem such as maximal curvature and noise level which are not known a priori in practice. Thus, in practice, hyper... More >