Lecture 10.2: New research directions (Multimodal Machine Learning, Carnegie Mellon University)

Lecture 10.2: New research directions (Multimodal Machine Learning, Carnegie Mellon University)

Topics: Recent approaches in multimodal ML

----------------------------------------------------------------------------------------------------------------
Carnegie Mellon University 11-777 Multimodal Machine Learning, 2020 Fall
Website: https://cmu-multicomp-lab.github.io/m...
Instructor: Louis-Philippe Morency

Multimodal machine learning (MMML) is a vibrant multi-disciplinary research field which studies computational approaches for modeling heterogenous data from multiple modalities. The course presents fundamental mathematical concepts in machine learning and deep learning relevant to the five main challenges in multimodal machine learning: (1) multimodal representation learning, (2) translation & mapping, (3) modality alignment, (4) multimodal fusion and (5) co-learning. The course also discusses recent state-of-the-art models and applications of multimodal machine learning.