Generalization Bounds for Neural Networks via Approximate Description Length

Authors: Amit Daniely, Elad Granot

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We investigate the sample complexity of networks with bounds on the magnitude of its weights... To establish our results we develop a new technique to analyze the sample complexity of families H of predictors. We start by defining a new notion of a randomized approximate description of functions f : X Rd. We then show that if there is a way to approximately describe functions in a class H using d bits, then d ϵ2 examples suffices to guarantee uniform convergence. Namely, that the empirical loss of all the functions in the class is ϵ-close to the true loss. Finally, we develop a set of tools for calculating the approximate description length of classes of functions that can be presented as a composition of linear function classes and non-linear functions.
Researcher Affiliation Collaboration Amit Daniely Hebrew University and Google Research Tel-Aviv amit.daniely@mail.huji.ac.il Elad Granot Hebrew University elad.granot@mail.huji.ac.il
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any concrete access information (e.g., a specific repository link or an explicit statement of code release) for the methodology described.
Open Datasets No The paper is theoretical and does not conduct experiments involving datasets, thus no information about public dataset availability is present.
Dataset Splits No The paper is theoretical and does not describe experiments with data, so there is no mention of training/validation/test dataset splits.
Hardware Specification No The paper is theoretical and does not detail any computational experiments, thus no specific hardware details are provided.
Software Dependencies No The paper is theoretical and does not describe any software implementations or dependencies with version numbers.
Experiment Setup No The paper is theoretical and does not include details about an experimental setup, such as hyperparameter values or training configurations.