Reading Group

This reading group at Oxford Statistics covers area related to Bayesian (nonparametric) methods, network modelling, statistical machine learning, power-laws in empirical data and the analysis of large neural networks. Email François if you are interested in participating, but not in the group.

2024

Date	Time	Room	Paper/Topic	Presenter	Notes
11/01/2024	11:00		Neurips papers	Kia, Stefano, Valentin, François

List of potentially interesting papers to read

Neurips 2023

• Score-based Generative Models with Lévy Processes. E. Yoon et al. https://openreview.net/forum?id=0Wp3VHX0Gm
• A Bayesian Take on Gaussian Process Networks. E. Giudice, J. Kuipers, G. Moffa. https://arxiv.org/abs/2306.11380
• Thin and Deep Gaussian processes. D de Souza et al. https://arxiv.org/abs/2310.11527
• Generalized test utilities for long-tail performance in extreme multi-label classification. E. Schultheis et al. link

Large neural networks/NTK

• A convergence analysis of gradient descent for deep linear neural networks. Arora et al. ICLR 2019. https://openreview.net/forum?id=SkMQg3C5K7
• On the explicit role of initialization on the convergence and implicit bias of overparametrized linear networks. Min et al., ICML 2021 https://proceedings.mlr.press/v139/min21c.html

Dataset Shift, imbalanced/extreme classification

Networks

Bayesian nonparametrics, power-laws