Toward Storage-Aware Learning with Compressed Data An Empirical Exploratory Study on JPEG

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lee, Kichang, Kim, Songkuk, Park, JaeYeon, Ko, JeongGil
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915692074237952
author Lee, Kichang
Kim, Songkuk
Park, JaeYeon
Ko, JeongGil
author_facet Lee, Kichang
Kim, Songkuk
Park, JaeYeon
Ko, JeongGil
contents On-device machine learning is often constrained by limited storage, particularly in continuous data collection scenarios. This paper presents an empirical study on storage-aware learning, focusing on the trade-off between data quantity and quality via compression. We demonstrate that naive strategies, such as uniform data dropping or one-size-fits-all compression, are suboptimal. Our findings further reveal that data samples exhibit varying sensitivities to compression, supporting the feasibility of a sample-wise adaptive compression strategy. These insights provide a foundation for developing a new class of storage-aware learning systems. The primary contribution of this work is the systematic characterization of this under-explored challenge, offering valuable insights that advance the understanding of storage-aware learning.
format Preprint
id arxiv_https___arxiv_org_abs_2508_12833
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Toward Storage-Aware Learning with Compressed Data An Empirical Exploratory Study on JPEG
Lee, Kichang
Kim, Songkuk
Park, JaeYeon
Ko, JeongGil
Machine Learning
Artificial Intelligence
68Txx
I.2; I.4.2; E.4
On-device machine learning is often constrained by limited storage, particularly in continuous data collection scenarios. This paper presents an empirical study on storage-aware learning, focusing on the trade-off between data quantity and quality via compression. We demonstrate that naive strategies, such as uniform data dropping or one-size-fits-all compression, are suboptimal. Our findings further reveal that data samples exhibit varying sensitivities to compression, supporting the feasibility of a sample-wise adaptive compression strategy. These insights provide a foundation for developing a new class of storage-aware learning systems. The primary contribution of this work is the systematic characterization of this under-explored challenge, offering valuable insights that advance the understanding of storage-aware learning.
title Toward Storage-Aware Learning with Compressed Data An Empirical Exploratory Study on JPEG
topic Machine Learning
Artificial Intelligence
68Txx
I.2; I.4.2; E.4
url https://arxiv.org/abs/2508.12833