Heterogeneous data

Mitigating Modality Collapse in Multimodal VAEs via Impartial Optimization
We connect the problem of modality collapse with conflicting gradients, and leverage solutions from multitask learning to alleviate it.
Adrián Javaloy, Maryam Meghdadi, Isabel Valera
Cite Code Poster Slides Video arXiv thread Proceedings
Mitigating Modality Collapse in Multimodal VAEs via Impartial Optimization
Lipschitz standardization for multivariate learning
We propose a new preprocessing method for first-order optimized probabilistic models that eases a more balanced learning across variables when modeling heterogeneous multivariate likelihoods.
Adrián Javaloy, Isabel Valera
PDF Cite Code arXiv
Lipschitz standardization for multivariate learning