Identifiable Deep Generative Models via Sparse Decoding

15 Jun 2022, 12:12 (modified: 20 Oct 2022, 17:53)Accepted by TMLREveryoneRevisionsBibTeX
Abstract: We develop the sparse VAE for unsupervised representation learning on high-dimensional data. The sparse VAE learns a set of latent factors (representations) which summarize the associations in the observed data features. The underlying model is sparse in that each observed feature (i.e. each dimension of the data) depends on a small subset of the latent factors. As examples, in ratings data each movie is only described by a few genres; in text data each word is only applicable to a few topics; in genomics, each gene is active in only a few biological processes. We prove such sparse deep generative models are identifiable: with infinite data, the true model parameters can be learned. (In contrast, most deep generative models are not identifiable.) We empirically study the sparse VAE with both simulated and real data. We find that it recovers meaningful latent factors and has smaller heldout reconstruction error than related methods.
License: Creative Commons Attribution 4.0 International (CC BY 4.0)
Submission Length: Regular submission (no more than 12 pages of main content)
Changes Since Last Submission: Camera ready revision - fixed eqs. 5 and 51 from conditional to joint
Assigned Action Editor: ~Andriy_Mnih1
Submission Number: 182