scientific article; zbMATH DE number 7626742

MaRDI QIDQ5053235zbMATH OpenFDO

Authors M. Pérez-Ortiz, Omar Risvaplata, John Shawe-Taylor, Csaba Szepesvári

Publication date 6 December 2022

Full work available at URL https://arxiv.org/abs/2007.12911, https://jmlr.csail.mit.edu/papers/v22/20-879.html

zbMATH Keywords

data-dependent priorsdeep learninggeneralisationneural work trainingPAC-Bayes with backproppathwise reparametrised gradientsweight randomisation

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Cites work

• A General Framework for Updating Belief Distributions
• A strongly quasiconvex PAC-Bayesian bound
• A useful theorem for nonlinear devices having Gaussian inputs
• Concentration inequalities. A nonasymptotic theory of independence
• Deep learning
• Emergence of invariance and disentanglement in deep representations
• scientific article; zbMATH DE number 6378127 (Why is no real title available?)
• scientific article; zbMATH DE number 7219345 (Why is no real title available?)
• scientific article; zbMATH DE number 5544465 (Why is no real title available?)
• scientific article; zbMATH DE number 149062 (Why is no real title available?)
• scientific article; zbMATH DE number 1222283 (Why is no real title available?)
• scientific article; zbMATH DE number 7272386 (Why is no real title available?)
• Microchoice bounds and self bounding learning algorithms
• Monte Carlo gradient estimation in machine learning
• Neural networks and deep learning. A textbook
• On the properties of variational approximations of Gibbs posteriors
• PAC-Bayes bounds with data dependent priors
• PAC-Bayesian analysis of co-clustering and beyond
• PAC-Bayesian Generalisation Error Bounds for Gaussian Process Classification
• PAC-Bayesian stochastic model selection
• Preserving statistical validity in adaptive data analysis (extended abstract)
• Risk bounds for the majority vote: from a PAC-Bayesian analysis to a learning algorithm
• Robust algorithms via PAC-Bayes and Laplace distributions
• Support Vector Machines
• The reusable holdout: preserving validity in adaptive data analysis
• Tighter PAC-Bayes bounds through distribution-dependent priors
• Understanding machine learning. From theory to algorithms

Cited in

(11)

• Sim-to-lab-to-real: safe reinforcement learning with shielding and generalization guarantees
• scientific article; zbMATH DE number 7306919 (Why is no real title available?)
• User-friendly Introduction to PAC-Bayes Bounds
• PACMAN: PAC-style bounds accounting for the mismatch between accuracy and negative log-loss
• Minimax optimality of deep neural networks on dependent data via PAC-Bayes bounds
• Tight PAC-Bayesian risk certificates for contrastive learning
• Operator theory, kernels, and feedforward neural networks
• Information-theoretic generalization bounds for transductive learning and its applications
• Wide stochastic networks: Gaussian limit and PAC-Bayesian training
• Generalization bounds: perspectives from information theory and PAC-Bayes
• Learning-to-optimize with PAC-Bayesian guarantees: theoretical considerations and practical implementation

This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5053235)