SCHA-VAE: Hierarchical Context Aggregation for Few-Shot Generation

Giorgio Giannone, Ole Winther
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:7550-7569, 2022.

Abstract

A few-shot generative model should be able to generate data from a novel distribution by only observing a limited set of examples. In few-shot learning the model is trained on data from many sets from distributions sharing some underlying properties such as sets of characters from different alphabets or objects from different categories. We extend current latent variable models for sets to a fully hierarchical approach with an attention-based point to set-level aggregation and call our method SCHA-VAE for Set-Context-Hierarchical-Aggregation Variational Autoencoder. We explore likelihood-based model comparison, iterative data sampling, and adaptation-free out-of-distribution generalization. Our results show that the hierarchical formulation better captures the intrinsic variability within the sets in the small data regime. This work generalizes deep latent variable approaches to few-shot learning, taking a step toward large-scale few-shot generation with a formulation that readily works with current state-of-the-art deep generative models.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-giannone22a, title = {{SCHA}-{VAE}: Hierarchical Context Aggregation for Few-Shot Generation}, author = {Giannone, Giorgio and Winther, Ole}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {7550--7569}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/giannone22a/giannone22a.pdf}, url = {https://proceedings.mlr.press/v162/giannone22a.html}, abstract = {A few-shot generative model should be able to generate data from a novel distribution by only observing a limited set of examples. In few-shot learning the model is trained on data from many sets from distributions sharing some underlying properties such as sets of characters from different alphabets or objects from different categories. We extend current latent variable models for sets to a fully hierarchical approach with an attention-based point to set-level aggregation and call our method SCHA-VAE for Set-Context-Hierarchical-Aggregation Variational Autoencoder. We explore likelihood-based model comparison, iterative data sampling, and adaptation-free out-of-distribution generalization. Our results show that the hierarchical formulation better captures the intrinsic variability within the sets in the small data regime. This work generalizes deep latent variable approaches to few-shot learning, taking a step toward large-scale few-shot generation with a formulation that readily works with current state-of-the-art deep generative models.} }
Endnote
%0 Conference Paper %T SCHA-VAE: Hierarchical Context Aggregation for Few-Shot Generation %A Giorgio Giannone %A Ole Winther %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-giannone22a %I PMLR %P 7550--7569 %U https://proceedings.mlr.press/v162/giannone22a.html %V 162 %X A few-shot generative model should be able to generate data from a novel distribution by only observing a limited set of examples. In few-shot learning the model is trained on data from many sets from distributions sharing some underlying properties such as sets of characters from different alphabets or objects from different categories. We extend current latent variable models for sets to a fully hierarchical approach with an attention-based point to set-level aggregation and call our method SCHA-VAE for Set-Context-Hierarchical-Aggregation Variational Autoencoder. We explore likelihood-based model comparison, iterative data sampling, and adaptation-free out-of-distribution generalization. Our results show that the hierarchical formulation better captures the intrinsic variability within the sets in the small data regime. This work generalizes deep latent variable approaches to few-shot learning, taking a step toward large-scale few-shot generation with a formulation that readily works with current state-of-the-art deep generative models.
APA
Giannone, G. & Winther, O.. (2022). SCHA-VAE: Hierarchical Context Aggregation for Few-Shot Generation. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:7550-7569 Available from https://proceedings.mlr.press/v162/giannone22a.html.

Related Material