Salvato in:
Dettagli Bibliografici
Autori principali: Reuter, Arik, Thielmann, Anton, Saefken, Benjamin
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2405.02295
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913341015851008
author Reuter, Arik
Thielmann, Anton
Saefken, Benjamin
author_facet Reuter, Arik
Thielmann, Anton
Saefken, Benjamin
contents Understanding how images influence the world, interpreting which effects their semantics have on various quantities and exploring the reasons behind changes in image-based predictions are highly difficult yet extremely interesting problems. By adopting a holistic modeling approach utilizing Neural Additive Models in combination with Diffusion Autoencoders, we can effectively identify the latent hidden semantics of image effects and achieve full intelligibility of additional tabular effects. Our approach offers a high degree of flexibility, empowering us to comprehensively explore the impact of various image characteristics. We demonstrate that the proposed method can precisely identify complex image effects in an ablation study. To further showcase the practical applicability of our proposed model, we conduct a case study in which we investigate how the distinctive features and attributes captured within host images exert influence on the pricing of Airbnb rentals.
format Preprint
id arxiv_https___arxiv_org_abs_2405_02295
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Neural Additive Image Model: Interpretation through Interpolation
Reuter, Arik
Thielmann, Anton
Saefken, Benjamin
Computer Vision and Pattern Recognition
Machine Learning
Understanding how images influence the world, interpreting which effects their semantics have on various quantities and exploring the reasons behind changes in image-based predictions are highly difficult yet extremely interesting problems. By adopting a holistic modeling approach utilizing Neural Additive Models in combination with Diffusion Autoencoders, we can effectively identify the latent hidden semantics of image effects and achieve full intelligibility of additional tabular effects. Our approach offers a high degree of flexibility, empowering us to comprehensively explore the impact of various image characteristics. We demonstrate that the proposed method can precisely identify complex image effects in an ablation study. To further showcase the practical applicability of our proposed model, we conduct a case study in which we investigate how the distinctive features and attributes captured within host images exert influence on the pricing of Airbnb rentals.
title Neural Additive Image Model: Interpretation through Interpolation
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2405.02295