Saved in:
Bibliographic Details
Main Authors: Bazinet, Mathieu, Zantedeschi, Valentina, Germain, Pascal
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.17932
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909533236887552
author Bazinet, Mathieu
Zantedeschi, Valentina
Germain, Pascal
author_facet Bazinet, Mathieu
Zantedeschi, Valentina
Germain, Pascal
contents The sample compression theory provides generalization guarantees for predictors that can be fully defined using a subset of the training dataset and a (short) message string, generally defined as a binary sequence. Previous works provided generalization bounds for the zero-one loss, which is restrictive notably when applied to deep learning approaches. In this paper, we present a general framework for deriving new sample compression bounds that hold for real-valued unbounded losses. Using the Pick-To-Learn (P2L) meta-algorithm, which transforms the training method of any machine-learning predictor to yield sample-compressed predictors, we empirically demonstrate the tightness of the bounds and their versatility by evaluating them on random forests and multiple types of neural networks.
format Preprint
id arxiv_https___arxiv_org_abs_2409_17932
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Sample Compression Unleashed: New Generalization Bounds for Real Valued Losses
Bazinet, Mathieu
Zantedeschi, Valentina
Germain, Pascal
Machine Learning
The sample compression theory provides generalization guarantees for predictors that can be fully defined using a subset of the training dataset and a (short) message string, generally defined as a binary sequence. Previous works provided generalization bounds for the zero-one loss, which is restrictive notably when applied to deep learning approaches. In this paper, we present a general framework for deriving new sample compression bounds that hold for real-valued unbounded losses. Using the Pick-To-Learn (P2L) meta-algorithm, which transforms the training method of any machine-learning predictor to yield sample-compressed predictors, we empirically demonstrate the tightness of the bounds and their versatility by evaluating them on random forests and multiple types of neural networks.
title Sample Compression Unleashed: New Generalization Bounds for Real Valued Losses
topic Machine Learning
url https://arxiv.org/abs/2409.17932