Hierarchical Few-Shot Generative Models
release_acgndbfvlvhadmk4r4be5ntma4
by
Giorgio Giannone, Ole Winther
2022
Abstract
A few-shot generative model should be able to generate data from a
distribution by only observing a limited set of examples. In few-shot learning
the model is trained on data from many sets from different distributions
sharing some underlying properties such as sets of characters from different
alphabets or sets of images of different type objects. 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 approach SCHA-VAE
for Set-Context-Hierarchical-Aggregation Variational Autoencoder. We explore
iterative data sampling, likelihood-based model comparison, 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. With this work we generalize deep latent variable approaches
to few-shot learning, taking a step towards large-scale few-shot generation
with a formulation that readily can work with current state-of-the-art deep
generative models.
In text/plain
format
Archived Files and Locations
application/pdf 2.5 MB
file_dexxgpewdrd75j6hpctjlv75pa
|
arxiv.org (repository) web.archive.org (webarchive) |
2110.12279v2
access all versions, variants, and formats of this works (eg, pre-prints)