Generalization bounds: Perspectives from information theory and PAC-Bayes

F Hellström, G Durisi, B Guedj… - … and Trends® in …, 2025 - nowpublishers.com
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 …

Sample-conditioned hypothesis stability sharpens information-theoretic generalization bounds

Z Wang, Y Mao - Advances in Neural Information …, 2024 - proceedings.neurips.cc
We present new information-theoretic generalization guarantees through the a novel
construction of the" neighboring-hypothesis" matrix and a new family of stability notions …

Minimum description length and generalization guarantees for representation learning

M Sefidgaran, A Zaidi… - Advances in Neural …, 2024 - proceedings.neurips.cc
A major challenge in designing efficient statistical supervised learning algorithms is finding
representations that perform well not only on available training samples but also on unseen …

Information-theoretic generalization bounds for learning from quantum data

MC Caro, T Gur, C Rouzé, DS Franca… - The Thirty Seventh …, 2024 - proceedings.mlr.press
Learning tasks play an increasingly prominent role in quantum information and computation.
They range from fundamental problems such as state discrimination and metrology over the …

Tighter information-theoretic generalization bounds from supersamples

Z Wang, Y Mao - arxiv preprint arxiv:2302.02432, 2023 - arxiv.org
In this work, we present a variety of novel information-theoretic generalization bounds for
learning algorithms, from the supersample setting of Steinke & Zakynthinou (2020)-the …

An information-theoretic approach to generalization theory

B Rodríguez-Gálvez, R Thobaben… - arxiv preprint arxiv …, 2024 - arxiv.org
We investigate the in-distribution generalization of machine learning algorithms. We depart
from traditional complexity-based approaches by analyzing information-theoretic bounds …

More PAC-Bayes bounds: From bounded losses, to losses with general tail behaviors, to anytime-validity

B Rodríguez-Gálvez, R Thobaben… - arxiv preprint arxiv …, 2023 - arxiv.org
In this paper, we present new high-probability PAC-Bayes bounds for different types of
losses. Firstly, for losses with a bounded range, we present a strengthened version of …

More PAC-Bayes bounds: From bounded losses, to losses with general tail behaviors, to anytime validity

B Rodríguez-Gálvez, R Thobaben… - Journal of Machine …, 2024 - jmlr.org
In this paper, we present new high-probability PAC-Bayes bounds for different types of
losses. Firstly, for losses with a bounded range, we recover a strengthened version of …

Information-Theoretic Generalization Bounds for Transductive Learning and its Applications

H Tang, Y Liu - arxiv preprint arxiv:2311.04561, 2023 - arxiv.org
In this paper, we develop data-dependent and algorithm-dependent generalization bounds
for transductive learning algorithms in the context of information theory for the first time. We …

Comparing Comparators in Generalization Bounds

F Hellström, B Guedj - International Conference on Artificial …, 2024 - proceedings.mlr.press
We derive generic information-theoretic and PAC-Bayesian generalization bounds involving
an arbitrary convex comparator function, which measures the discrepancy between the …