Multimodal learning: Difference between revisions

Revision as of 21:25, 10 November 2022

Revision as of 11:24, 11 November 2022

Line 6:

}}

<!--- Don't mess with this line! ---><!--- Write your article below this line --->

'''Multimodal learning''' attempts to model the combination of different [[Modality (human–computer interaction)|modalities]] of data, often rising in real-world applications. ExampleAn ofexample suchof joint data is combining text, (typically represented as discrete word count vectors,) with imaging data comprising of [[pixel]] intensities and annotation tags. As these modalities have fundamentally different statistical properties, combining them is non-trivial, andwhich is suchwhy specialized modelling strategies and algorithms are required.

==Motivation==