Identifiability of deep generative models without auxiliary information

Authors: Bohdan Kivva, Goutham Rajendran, Pradeep Ravikumar, Bryon Aragam

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our main result is an identifiability hierarchy that significantly generalizes previous work and exposes how different assumptions lead to different strengths of identifiability, and includes certain vanilla VAEs with isotropic Gaussian priors as a special case. For example, our weakest result establishes (unsupervised) identifiability up to an affine transformation, and thus partially resolves an open problem regarding model identifiability raised in prior work. These theoretical results are augmented with experiments on both simulated and real data.
Researcher Affiliation Academia Bohdan Kivva University of Chicago bkivva@uchicago.edu Goutham Rajendran University of Chicago goutham@uchicago.edu Pradeep Ravikumar Carnegie Mellon University pradeepr@cs.cmu.edu Bryon Aragam University of Chicago bryon@chicagobooth.edu
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks in the main text.
Open Source Code Yes 3. If you ran experiments... (a) Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [Yes]
Open Datasets Yes Real data We measure stability of the learnt latent space by training MFCVAE [18] on MNIST 10 times with different initializations and then comparing the latent representations learnt.
Dataset Splits Yes 3. If you ran experiments... (b) Did you specify all the training details (e.g., data splits, hyperparameters, how they were chosen)? [Yes]
Hardware Specification No 3. If you ran experiments... (d) Did you include the total amount of compute and the type of resources used (e.g., type of GPUs, internal cluster, or cloud provider)? [Yes] See supplement.
Software Dependencies No The paper does not provide specific software dependencies with version numbers in the main text.
Experiment Setup No Definitions of these metrics and additional details on the experiments can be found in Appendix J.