Quantitative CLTs in Deep Neural Networks Date: 16 November 2023, Thursday Time: 6:30 pm AEDT Statistics and Stochastic colloquium (part of the Colloquium Series of the Department of Mathematics and Statistics) at La Trobe University jointly organized with the Probability Victoria Seminar. Contact the organizers: Andriy Olenko a.olenko@latrobe.edu.au, Kostya Borovkov kostya.borovkov@gmail.com Speaker: Ivan Nourdin (Universitéit Lëtzebuerg, Grand Duchy of Luxembourg) Abstract: In this talk, we will study the distribution of a fully connected neural network with random Gaussian weights and biases in which the hidden layer widths are proportional to a large constant n . More precisely, we will explain how to prove quantitative bounds on normal approximations valid at large but finite n and any fixed network depth. This is based on a joint work with S. Favaro, B. Hanin, D. Marinucci and G. Peccati. Zoom meeting link: https://unimelb.zoom.us/j/82073536928?pwd=b3dmOUFwdFNUZS9hMWxHZk
18 August, 4 PM AET Luke Yates Postdoctoral Research Fellow, University of Tasmania Hybrid: Anita B. Lawrence Centre 4082 Zoom link: https://unsw.zoom.us/j/88495626621 Title: New tools for time series analysis of 'omics' data Abstract: It is peak hour in downtown Transcriptome, where millions of messenger RNA are en route from their local DNA carrying important instructions for protein synthesis and other civic services. A typical transcriptomics (RNA-seq) data set is a snapshot of this busy scene, comprising a sample of extracts of these instruction sequences. A key goal in molecular biology is to determine which sequences (i.e., sets of transcripts or genes) change their expression in response to different treatment conditions to discover molecular mechanisms for biological traits. In this talk, I will give a brief background on the sampling, technical, and statistical processes involved in generating such data sets before focusing on new and existing methods for their