Date/Time : 14 Feb 2025 10:00
Venue : S17 #05-11 (SR5)
Speaker : Luca Makowiec, National University of Singapore

We will introduce Random Spanning Trees in Random Environment (RSTRE), a disordered system on spanning trees that interpolates between the Uniform Spanning Tree (UST) and Minimum Spanning Tree (MST) measures. Our primary goal is to study how local and global observables of the RSTRE depend on the inverse temperature beta. Of particular interest is the diameter of a (typical) spanning tree, as it is the first step towards establishing convergence to a non-trivial scaling limit. Furthermore, for the RSTRE on the complete graph with uniform disorder, we discuss a sharp transition of the local limit of the RSTRE to either the UST or MST local limit. This stands in contrast to our conjectured smooth transition of the diameter when beta is inside the so-called intermediate regime.

Add to calendar :

Date/Time : 28 Nov 2024 13:30
Venue : S17 #04-05 (SR2)
Speaker : Peter Buhlmann, ETH Zurich

Nonparametric function estimation and prediction with moderate or large dimension of the covariates are particularly susceptible to extrapolation, because data points are typically far apart from each other in such moderate or higher dimension. Thus, there is a need to have machine learning methods that are extrapolation-aware, i.e. that automatically perform well (in a sense) when extrapolation occurs. Without such extrapolation-aware techniques, inference from standard machine learning and nonparametric procedures may be poor or invalid. We introduce a novel conceptual framework and introduce *Xtrapolation* which allows for extrapolation-aware inference with any ML algorithm.

This is joint work with Niklas Pfister (University of Copenhagen)

Add to calendar :

Date/Time : 01 Nov 2024 15:00
Venue : S17 #04-06 (SR1)
Speaker : Nathan Ross, University of Melbourne

Neural networks are random parametric functions, where the parameters are chosen according to stochastic gradient descent, based on the sample of training data. It has long been known that at initialisation, a neural network with “large” width and “fixed” depth can be well-approximated by a limiting Gaussian random field. But what quantities are well-approximated, and how wide does a neural network need to be for the approximation to be good? There has been a flurry of recent work trying to answer these and related questions, by deriving rates of convergence between the distribution of the neural network and the limiting Gaussian random field in various probability metrics, including some at the process level.

The aim of these lectures is to explain a general approach that can be applied to derive such rates at the process level, which uses Stein’s method, functional smoothing inequalities, and classical Gaussian process concepts such as Karhunen-Loeve expansion and Cameron-Martin/Reproducing Kernel Hilbert Space. The target audience is PhD students planning to do research in probability or statistics. In particular, the definitions and relevant properties of the concepts above will be covered in detail. Open questions and future directions will also be discussed.

The approach is based on joint work with (A.D. Barbour, K. Balasubramanian, L. Goldstein, A. Salim, and G. Zheng).

Add to calendar :

Date/Time : 30 Oct 2024 16:00
Venue : S17 #04-05 (SR2)
Speaker : Jimmy He, Ohio State University

The study of quantum groups and integrable systems have long been connected through the Yang-Baxter equation. Recently, these ideas have been introduced in probability to study integrable stochastic models. I will discuss a similar story for models with boundary, which are connected to quantum symmetric spaces through the reflection equation, as well as some applications to probability. The focus will be on illustrating the connections through illustrative examples and applications for a specific model called the six vertex model.

Add to calendar :

Date/Time : 30 Oct 2024 15:00
Venue : S17 #04-04 (SR3)
Speaker : Nathan Ross, University of Melbourne

Neural networks are random parametric functions, where the parameters are chosen according to stochastic gradient descent, based on the sample of training data. It has long been known that at initialisation, a neural network with “large” width and “fixed” depth can be well-approximated by a limiting Gaussian random field. But what quantities are well-approximated, and how wide does a neural network need to be for the approximation to be good? There has been a flurry of recent work trying to answer these and related questions, by deriving rates of convergence between the distribution of the neural network and the limiting Gaussian random field in various probability metrics, including some at the process level.

The aim of these lectures is to explain a general approach that can be applied to derive such rates at the process level, which uses Stein’s method, functional smoothing inequalities, and classical Gaussian process concepts such as Karhunen-Loeve expansion and Cameron-Martin/Reproducing Kernel Hilbert Space. The target audience is PhD students planning to do research in probability or statistics. In particular, the definitions and relevant properties of the concepts above will be covered in detail. Open questions and future directions will also be discussed.

The approach is based on joint work with (A.D. Barbour, K. Balasubramanian, L. Goldstein, A. Salim, and G. Zheng).

Add to calendar :

Date/Time : 29 Oct 2024 14:00
Venue : S17 #04-06 (SR1)
Speaker : Nathan Ross, University of Melbourne

Neural networks are random parametric functions, where the parameters are chosen according to stochastic gradient descent, based on the sample of training data. It has long been known that at initialisation, a neural network with “large” width and “fixed” depth can be well-approximated by a limiting Gaussian random field. But what quantities are well-approximated, and how wide does a neural network need to be for the approximation to be good? There has been a flurry of recent work trying to answer these and related questions, by deriving rates of convergence between the distribution of the neural network and the limiting Gaussian random field in various probability metrics, including some at the process level.

The aim of these lectures is to explain a general approach that can be applied to derive such rates at the process level, which uses Stein’s method, functional smoothing inequalities, and classical Gaussian process concepts such as Karhunen-Loeve expansion and Cameron-Martin/Reproducing Kernel Hilbert Space. The target audience is PhD students planning to do research in probability or statistics. In particular, the definitions and relevant properties of the concepts above will be covered in detail. Open questions and future directions will also be discussed.

The approach is based on joint work with (A.D. Barbour, K. Balasubramanian, L. Goldstein, A. Salim, and G. Zheng).

Add to calendar :

Date/Time : 28 Oct 2024 16:00
Venue : S17 #04-06 (SR1)
Speaker : Nathan Ross, University of Melbourne

Neural networks are random parametric functions, where the parameters are chosen according to stochastic gradient descent, based on the sample of training data. It has long been known that at initialisation, a neural network with “large” width and “fixed” depth can be well-approximated by a limiting Gaussian random field. But what quantities are well-approximated, and how wide does a neural network need to be for the approximation to be good? There has been a flurry of recent work trying to answer these and related questions, by deriving rates of convergence between the distribution of the neural network and the limiting Gaussian random field in various probability metrics, including some at the process level.

The aim of these lectures is to explain a general approach that can be applied to derive such rates at the process level, which uses Stein’s method, functional smoothing inequalities, and classical Gaussian process concepts such as Karhunen-Loeve expansion and Cameron-Martin/Reproducing Kernel Hilbert Space. The target audience is PhD students planning to do research in probability or statistics. In particular, the definitions and relevant properties of the concepts above will be covered in detail. Open questions and future directions will also be discussed.

The approach is based on joint work with (A.D. Barbour, K. Balasubramanian, L. Goldstein, A. Salim, and G. Zheng).

Add to calendar :

Date/Time : 24 Oct 2024 17:00
Venue : S17 #05-12 (SR4)
Speaker : Jimmy He, Ohio State University

The study of quantum groups and integrable systems have long been connected through the Yang-Baxter equation. Recently, these ideas have been introduced in probability to study integrable stochastic models. I will discuss a similar story for models with boundary, which are connected to quantum symmetric spaces through the reflection equation, as well as some applications to probability. The focus will be on illustrating the connections through illustrative examples and applications for a specific model called the six vertex model.

Add to calendar :

Date/Time : 23 Oct 2024 16:00
Venue : S17 #04-05 (SR2)
Speaker : Jimmy He, Ohio State University

The study of quantum groups and integrable systems have long been connected through the Yang-Baxter equation. Recently, these ideas have been introduced in probability to study integrable stochastic models. I will discuss a similar story for models with boundary, which are connected to quantum symmetric spaces through the reflection equation, as well as some applications to probability. The focus will be on illustrating the connections through illustrative examples and applications for a specific model called the six vertex model.

Add to calendar :