Machine Learning – Publications
1.
Javaloy, Adrián; Meghdadi, Maryam; Valera, Isabel
Mitigating Modality Collapse in Multimodal VAEs via Impartial Optimization Journal Article Spotlight
In: CoRR, vol. abs/2206.04496, 2022.
Abstract | Links | BibTeX | Tags: adrian, isabel, maryam, project-robustgenerative, spotlight, variational autoencoder
@article{DBLP:journals/corr/abs-2206-04496,
title = {Mitigating Modality Collapse in Multimodal VAEs via Impartial Optimization},
author = {Adrián Javaloy and Maryam Meghdadi and Isabel Valera},
url = {https://doi.org/10.48550/arXiv.2206.04496},
doi = {10.48550/arXiv.2206.04496},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
journal = {CoRR},
volume = {abs/2206.04496},
abstract = {A number of variational autoencoders (VAEs) have recently emerged with the aim of modeling multimodal data, e.g., to jointly model images and their corresponding captions. Still, multimodal VAEs tend to focus solely on a subset of the modalities, e.g., by fitting the image while neglecting the caption. We refer to this limitation as modality collapse. In this work, we argue that this effect is a consequence of conflicting gradients during multimodal VAE training. We show how to detect the sub-graphs in the computational graphs where gradients conflict (impartiality blocks), as well as how to leverage existing gradient-conflict solutions from multitask learning to mitigate modality collapse. That is, to ensure impartial optimization across modalities. We apply our training framework to several multimodal VAE models, losses and datasets from the literature, and empirically show that our framework significantly improves the reconstruction performance, conditional generation, and coherence of the latent space across modalities.},
keywords = {adrian, isabel, maryam, project-robustgenerative, spotlight, variational autoencoder},
pubstate = {published},
tppubtype = {article}
}
A number of variational autoencoders (VAEs) have recently emerged with the aim of modeling multimodal data, e.g., to jointly model images and their corresponding captions. Still, multimodal VAEs tend to focus solely on a subset of the modalities, e.g., by fitting the image while neglecting the caption. We refer to this limitation as modality collapse. In this work, we argue that this effect is a consequence of conflicting gradients during multimodal VAE training. We show how to detect the sub-graphs in the computational graphs where gradients conflict (impartiality blocks), as well as how to leverage existing gradient-conflict solutions from multitask learning to mitigate modality collapse. That is, to ensure impartial optimization across modalities. We apply our training framework to several multimodal VAE models, losses and datasets from the literature, and empirically show that our framework significantly improves the reconstruction performance, conditional generation, and coherence of the latent space across modalities.
2.
Javaloy, Adrián; Meghdadi, Maryam; Valera, Isabel
Boosting heterogeneous VAEs via multi-objective optimization Workshop
2021.
Abstract | Links | BibTeX | Tags: adrian, isabel, maryam, project-robustgenerative
@workshop{nokey,
title = {Boosting heterogeneous VAEs via multi-objective optimization},
author = {Adrián Javaloy and Maryam Meghdadi and Isabel Valera},
url = {http://adrian.javaloy.com/publication/mo-vae/},
year = {2021},
date = {2021-12-01},
urldate = {2021-12-01},
abstract = {A number of variational autoencoders (VAEs) have recently emerged with the aim of modeling multimodal data, e.g., to jointly model images and their corresponding captions. Still, multimodal VAEs tend to focus solely on a subset of the modalities, e.g., by fitting the image while neglecting the caption. We refer to this limitation as modality collapse. In this work, we argue that this effect is a consequence of conflicting gradients during multimodal VAE training. We show how to detect the sub-graphs in the computational graphs where gradients conflict (impartiality blocks), as well as how to leverage existing gradient-conflict solutions from multitask learning to mitigate modality collapse. That is, to ensure impartial optimization across modalities. We apply our training framework to several multimodal VAE models, losses and datasets from the literature, and empirically show that our framework significantly improves the reconstruction performance, conditional generation, and coherence of the latent space across modalities.},
keywords = {adrian, isabel, maryam, project-robustgenerative},
pubstate = {published},
tppubtype = {workshop}
}
A number of variational autoencoders (VAEs) have recently emerged with the aim of modeling multimodal data, e.g., to jointly model images and their corresponding captions. Still, multimodal VAEs tend to focus solely on a subset of the modalities, e.g., by fitting the image while neglecting the caption. We refer to this limitation as modality collapse. In this work, we argue that this effect is a consequence of conflicting gradients during multimodal VAE training. We show how to detect the sub-graphs in the computational graphs where gradients conflict (impartiality blocks), as well as how to leverage existing gradient-conflict solutions from multitask learning to mitigate modality collapse. That is, to ensure impartial optimization across modalities. We apply our training framework to several multimodal VAE models, losses and datasets from the literature, and empirically show that our framework significantly improves the reconstruction performance, conditional generation, and coherence of the latent space across modalities.