Beyond Unimodal Learning: The Importance of Integrating Multiple Modalities for Lifelong Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sarfraz, Fahad, Zonooz, Bahram, Arani, Elahe
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917657262948352
author Sarfraz, Fahad
Zonooz, Bahram
Arani, Elahe
author_facet Sarfraz, Fahad
Zonooz, Bahram
Arani, Elahe
contents While humans excel at continual learning (CL), deep neural networks (DNNs) exhibit catastrophic forgetting. A salient feature of the brain that allows effective CL is that it utilizes multiple modalities for learning and inference, which is underexplored in DNNs. Therefore, we study the role and interactions of multiple modalities in mitigating forgetting and introduce a benchmark for multimodal continual learning. Our findings demonstrate that leveraging multiple views and complementary information from multiple modalities enables the model to learn more accurate and robust representations. This makes the model less vulnerable to modality-specific regularities and considerably mitigates forgetting. Furthermore, we observe that individual modalities exhibit varying degrees of robustness to distribution shift. Finally, we propose a method for integrating and aligning the information from different modalities by utilizing the relational structural similarities between the data points in each modality. Our method sets a strong baseline that enables both single- and multimodal inference. Our study provides a promising case for further exploring the role of multiple modalities in enabling CL and provides a standard benchmark for future research.
format Preprint
id arxiv_https___arxiv_org_abs_2405_02766
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Beyond Unimodal Learning: The Importance of Integrating Multiple Modalities for Lifelong Learning
Sarfraz, Fahad
Zonooz, Bahram
Arani, Elahe
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
While humans excel at continual learning (CL), deep neural networks (DNNs) exhibit catastrophic forgetting. A salient feature of the brain that allows effective CL is that it utilizes multiple modalities for learning and inference, which is underexplored in DNNs. Therefore, we study the role and interactions of multiple modalities in mitigating forgetting and introduce a benchmark for multimodal continual learning. Our findings demonstrate that leveraging multiple views and complementary information from multiple modalities enables the model to learn more accurate and robust representations. This makes the model less vulnerable to modality-specific regularities and considerably mitigates forgetting. Furthermore, we observe that individual modalities exhibit varying degrees of robustness to distribution shift. Finally, we propose a method for integrating and aligning the information from different modalities by utilizing the relational structural similarities between the data points in each modality. Our method sets a strong baseline that enables both single- and multimodal inference. Our study provides a promising case for further exploring the role of multiple modalities in enabling CL and provides a standard benchmark for future research.
title Beyond Unimodal Learning: The Importance of Integrating Multiple Modalities for Lifelong Learning
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.02766