Robust Multimodal Learning with Missing Modalities via Parameter-Efficient Adaptation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Reza, Md Kaykobad, Prater-Bennette, Ashley, Asif, M. Salman
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929536531169280
author Reza, Md Kaykobad
Prater-Bennette, Ashley
Asif, M. Salman
author_facet Reza, Md Kaykobad
Prater-Bennette, Ashley
Asif, M. Salman
contents Multimodal learning seeks to utilize data from multiple sources to improve the overall performance of downstream tasks. It is desirable for redundancies in the data to make multimodal systems robust to missing or corrupted observations in some correlated modalities. However, we observe that the performance of several existing multimodal networks significantly deteriorates if one or multiple modalities are absent at test time. To enable robustness to missing modalities, we propose a simple and parameter-efficient adaptation procedure for pretrained multimodal networks. In particular, we exploit modulation of intermediate features to compensate for the missing modalities. We demonstrate that such adaptation can partially bridge performance drop due to missing modalities and outperform independent, dedicated networks trained for the available modality combinations in some cases. The proposed adaptation requires extremely small number of parameters (e.g., fewer than 1% of the total parameters) and applicable to a wide range of modality combinations and tasks. We conduct a series of experiments to highlight the missing modality robustness of our proposed method on five different multimodal tasks across seven datasets. Our proposed method demonstrates versatility across various tasks and datasets, and outperforms existing methods for robust multimodal learning with missing modalities.
format Preprint
id arxiv_https___arxiv_org_abs_2310_03986
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Robust Multimodal Learning with Missing Modalities via Parameter-Efficient Adaptation
Reza, Md Kaykobad
Prater-Bennette, Ashley
Asif, M. Salman
Computer Vision and Pattern Recognition
Machine Learning
Multimodal learning seeks to utilize data from multiple sources to improve the overall performance of downstream tasks. It is desirable for redundancies in the data to make multimodal systems robust to missing or corrupted observations in some correlated modalities. However, we observe that the performance of several existing multimodal networks significantly deteriorates if one or multiple modalities are absent at test time. To enable robustness to missing modalities, we propose a simple and parameter-efficient adaptation procedure for pretrained multimodal networks. In particular, we exploit modulation of intermediate features to compensate for the missing modalities. We demonstrate that such adaptation can partially bridge performance drop due to missing modalities and outperform independent, dedicated networks trained for the available modality combinations in some cases. The proposed adaptation requires extremely small number of parameters (e.g., fewer than 1% of the total parameters) and applicable to a wide range of modality combinations and tasks. We conduct a series of experiments to highlight the missing modality robustness of our proposed method on five different multimodal tasks across seven datasets. Our proposed method demonstrates versatility across various tasks and datasets, and outperforms existing methods for robust multimodal learning with missing modalities.
title Robust Multimodal Learning with Missing Modalities via Parameter-Efficient Adaptation
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2310.03986