Knowledge Distillation for a Domain-Adaptive Visual Recommender System

In the last few years large-scale foundational models have shown remarkable performance in computer vision tasks. However, deploying such models in a production environment poses a significant challenge, because of their computational requirements. Furthermore, these models typically produce generic...

Full description

Saved in:
Bibliographic Details
Main Authors: Alessandro Abluton, Luigi Portinale
Format: Article
Language:English
Published: LibraryPress@UF 2024-05-01
Series:Proceedings of the International Florida Artificial Intelligence Research Society Conference
Online Access:https://journals.flvc.org/FLAIRS/article/view/135533
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849736803328720896
author Alessandro Abluton
Luigi Portinale
author_facet Alessandro Abluton
Luigi Portinale
author_sort Alessandro Abluton
collection DOAJ
description In the last few years large-scale foundational models have shown remarkable performance in computer vision tasks. However, deploying such models in a production environment poses a significant challenge, because of their computational requirements. Furthermore, these models typically produce generic results and they often need some sort of external input. The concept of knowledge distillation provides a promising solution to this problem. In this paper, we focus on the challenges faced in the application of knowledge distillation techniques in the task of augmenting a dataset for object detection used in a commercial Visual Recommender System called VISIDEA; the goal consists in detecting items in various e-commerce websites, encompassing a wide range of custom product categories. We discuss a possible solution to problems such as label duplication, erroneous labeling and lack of robustness to prompting, by considering examples in the field of fashion apparel recommendation.
format Article
id doaj-art-14e8b1e800f14690aa0a9eee65ceb4ed
institution DOAJ
issn 2334-0754
2334-0762
language English
publishDate 2024-05-01
publisher LibraryPress@UF
record_format Article
series Proceedings of the International Florida Artificial Intelligence Research Society Conference
spelling doaj-art-14e8b1e800f14690aa0a9eee65ceb4ed2025-08-20T03:07:10ZengLibraryPress@UFProceedings of the International Florida Artificial Intelligence Research Society Conference2334-07542334-07622024-05-013710.32473/flairs.37.1.13553371912Knowledge Distillation for a Domain-Adaptive Visual Recommender SystemAlessandro Abluton0Luigi Portinale1https://orcid.org/0000-0002-6053-4667Inferendo srlUniversity of Eastern PiedmontIn the last few years large-scale foundational models have shown remarkable performance in computer vision tasks. However, deploying such models in a production environment poses a significant challenge, because of their computational requirements. Furthermore, these models typically produce generic results and they often need some sort of external input. The concept of knowledge distillation provides a promising solution to this problem. In this paper, we focus on the challenges faced in the application of knowledge distillation techniques in the task of augmenting a dataset for object detection used in a commercial Visual Recommender System called VISIDEA; the goal consists in detecting items in various e-commerce websites, encompassing a wide range of custom product categories. We discuss a possible solution to problems such as label duplication, erroneous labeling and lack of robustness to prompting, by considering examples in the field of fashion apparel recommendation.https://journals.flvc.org/FLAIRS/article/view/135533
spellingShingle Alessandro Abluton
Luigi Portinale
Knowledge Distillation for a Domain-Adaptive Visual Recommender System
Proceedings of the International Florida Artificial Intelligence Research Society Conference
title Knowledge Distillation for a Domain-Adaptive Visual Recommender System
title_full Knowledge Distillation for a Domain-Adaptive Visual Recommender System
title_fullStr Knowledge Distillation for a Domain-Adaptive Visual Recommender System
title_full_unstemmed Knowledge Distillation for a Domain-Adaptive Visual Recommender System
title_short Knowledge Distillation for a Domain-Adaptive Visual Recommender System
title_sort knowledge distillation for a domain adaptive visual recommender system
url https://journals.flvc.org/FLAIRS/article/view/135533
work_keys_str_mv AT alessandroabluton knowledgedistillationforadomainadaptivevisualrecommendersystem
AT luigiportinale knowledgedistillationforadomainadaptivevisualrecommendersystem