PAC-Bayes Analysis Beyond the Usual Bounds

Authors: Omar Rivasplata, Ilja Kuzborskij, Csaba Szepesvari, John Shawe-Taylor

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this paper we discuss a basic PAC-Bayes inequality (Theorem 1 below) and a general template for PAC-Bayesian bounds (Theorem 2 below). The formulation of both these results is based on representing data-dependent distributions as stochastic kernels.
Researcher Affiliation Collaboration Omar Rivasplata University College London & Deep Mind o.rivasplata@cs.ucl.ac.uk Ilja Kuzborskij Deep Mind iljak@google.com Csaba Szepesv ari Deep Mind szepi@google.com John Shawe-Taylor University College London jst@cs.ucl.ac.uk
Pseudocode No The paper does not contain any pseudocode or algorithm blocks. It presents theorems and mathematical proofs.
Open Source Code No The paper does not mention providing open-source code for the described methodology.
Open Datasets No This is a theoretical paper focusing on mathematical frameworks and theorems. It does not conduct empirical studies using specific datasets, so no dataset availability information is provided.
Dataset Splits No This is a theoretical paper and does not describe empirical experiments with training, validation, or test data splits.
Hardware Specification No This is a theoretical paper and does not describe experiments that would require hardware specifications.
Software Dependencies No This is a theoretical paper. No software dependencies with specific version numbers are mentioned as it does not describe empirical experiments.
Experiment Setup No This is a theoretical paper and does not describe empirical experiments with specific setup details or hyperparameters.