Aggregated predictors are obtained by making a set of basic predictors vote according to
some weights, that is, to some probability distribution. Randomized predictors are obtained …

A fundamental question in theoretical machine learning is generalization. Over the past
decades, the PAC-Bayesian approach has been established as a flexible framework to …

We present a unified framework for deriving PAC-Bayesian generalization bounds. Unlike
most previous literature on this topic, our bounds are anytime-valid (ie, time-uniform) …

Modern language models can contain billions of parameters, raising the question of whether
they can generalize beyond the training data or simply parrot their training corpora. We …

Large language models (LLMs) are trained on a deluge of text data with limited quality
control. As a result, LLMs can exhibit unintended or even harmful behaviours, such as …

While PAC-Bayes is now an established learning framework for light-tailed losses (\emph
{eg}, subgaussian or subexponential), its extension to the case of heavy-tailed losses …

We are motivated by the problem of providing strong generalization guarantees in the
context of meta-learning. Existing generalization bounds are either challenging to evaluate …

We present a new family of information-theoretic generalization bounds, in which the
training loss and the population loss are compared through a jointly convex function. This …

Minimising upper bounds on the population risk or the generalisation gap has been widely
used in structural risk minimisation (SRM)--this is in particular at the core of PAC-Bayesian …

In this paper, the worst-case probability measure over the data is introduced as a tool for
characterizing the generalization capabilities of machine learning algorithms. More …