Lossy Source Coding with Focal Loss

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Dytso, Alex, Cardone, Martina
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910920733622272
author Dytso, Alex
Cardone, Martina
author_facet Dytso, Alex
Cardone, Martina
contents Focal loss has recently gained significant popularity, particularly in tasks like object detection where it helps to address class imbalance by focusing more on hard-to-classify examples. This work proposes the focal loss as a distortion measure for lossy source coding. The paper provides single-shot converse and achievability bounds. These bounds are then used to characterize the distortion-rate trade-off in the infinite blocklength, which is shown to be the same as that for the log loss case. In the non-asymptotic case, the difference between focal loss and log loss is illustrated through a series of simulations.
format Preprint
id arxiv_https___arxiv_org_abs_2504_19913
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Lossy Source Coding with Focal Loss
Dytso, Alex
Cardone, Martina
Information Theory
Focal loss has recently gained significant popularity, particularly in tasks like object detection where it helps to address class imbalance by focusing more on hard-to-classify examples. This work proposes the focal loss as a distortion measure for lossy source coding. The paper provides single-shot converse and achievability bounds. These bounds are then used to characterize the distortion-rate trade-off in the infinite blocklength, which is shown to be the same as that for the log loss case. In the non-asymptotic case, the difference between focal loss and log loss is illustrated through a series of simulations.
title Lossy Source Coding with Focal Loss
topic Information Theory
url https://arxiv.org/abs/2504.19913