paper-with-me

홈 › Papers

Hyperprior Induced Unsupervised Disentanglement of Latent Representations

2018-09-12 · Abdul Fatir Ansari, Harold Soh

We address the problem of unsupervised disentanglement of latent representations learnt via deep generative models. In contrast to current approaches that operate on the evidence lower bound (ELBO), we argue that statistical independence in the latent space of VAEs can be enforced in a principled hierarchical Bayesian manner. To this effect, we augment the standard VAE with an inverse-Wishart (IW) prior on the covariance matrix of the latent code. By tuning the IW parameters, we are able to encourage (or discourage) independence in the learnt latent dimensions. Extensive experimental results on a range of datasets (2DShapes, 3DChairs, 3DFaces and CelebA) show our approach to outperform the $\beta$-VAE and is competitive with the state-of-the-art FactorVAE. Our approach achieves significantly better disentanglement and reconstruction on a new dataset (CorrelatedEllipses) which introduces correlations between the factors of variation.

📄 PDF Abstract BibTeX arXiv:1809.04497

Code (2)

crslab/CHyVAE 공식 구현 tf
crslab/correlated-ellipses

Tasks

Disentanglement

Methods 이 논문이 사용한 방법론

USD Coin Customer Service Number +1-833-534-1729 설명 없음

Similar Papers 제목 키워드 기반

Unsupervised Disentanglement with Tensor Product Representations on the Torus

2022-02-13 · ICLR 2022 4 · Michael Rotman, Amit Dekel, Shir Gur, Yaron Oz 외

The current methods for learning representations with auto-encoders almost exclusively employ vectors as the latent representations. In this work, we propose to employ a tensor product structure for this purpose. This wa…

DisentanglementInformativeness

Towards Unsupervised Content Disentanglement in Sentence Representations via Syntactic Roles

2022-06-22 · Ghazi Felhi, Joseph Le Roux, Djamé Seddah

Linking neural representations to linguistic factors is crucial in order to build and analyze NLP models interpretable by humans. Among these factors, syntactic roles (e.g. subjects, direct objects,$\dots$) and their rea…

DecoderDisentanglementMachine TranslationSentence

Where and What? Examining Interpretable Disentangled Representations

2021-04-07 · CVPR 2021 1 · Xinqi Zhu, Chang Xu, DaCheng Tao

Capturing interpretable variations has long been one of the goals in disentanglement learning. However, unlike the independence assumption, interpretability has rarely been exploited to encourage disentanglement in the u…

DisentanglementModel SelectionPerceptual Distance

Variational Inference of Disentangled Latent Concepts from Unlabeled Observations

2017-11-02 · ICLR 2018 1 · Abhishek Kumar, Prasanna Sattigeri, Avinash Balakrishnan

Disentangled representations, where the higher level data generative factors are reflected in disjoint latent dimensions, offer several benefits such as ease of deriving invariant representations, transferability to othe…

DisentanglementVariational Inference

On Disentangled Representations Learned From Correlated Data

2020-06-14 · Frederik Träuble, Elliot Creager, Niki Kilbertus, Francesco Locatello 외

The focus of disentanglement approaches has been on identifying independent factors of variation in data. However, the causal variables underlying real-world observations are often not statistically independent. In this …

DisentanglementFairness