A Novel Variational Autoencoder with Applications to Generative Modelling, Classification, and Ordinal Regression

Jaskari, Joel; Kivinen, Jyri J.

doi:10.48550/arxiv.1812.07352

Cited by 2 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[6] assumes ordinal paired data instances are available in the training data, which is different from the ordinal classification task [42]. The variational posterior of the ordinal class is introduced in [22]. [15] utilizes the video-level label in the target domain and does not require the extracted latent vectors to be aligned with the ordinal constraints.…”

Section: Related Workmentioning

confidence: 99%

Recursively Conditional Gaussian for Ordinal Unsupervised Domain Adaptation

Liu

et al. 2021

Preprint

View full text Add to dashboard Cite

There has been a growing interest in unsupervised domain adaptation (UDA) to alleviate the data scalability issue, while the existing works usually focus on classifying independently discrete labels. However, in many tasks (e.g., medical diagnosis), the labels are discrete and successively distributed. The UDA for ordinal classification requires inducing non-trivial ordinal distribution prior to the latent space. Target for this, the partially ordered set (poset) is defined for constraining the latent vector. Instead of the typically i.i.d. Gaussian latent prior, in this work, a recursively conditional Gaussian (RCG) set is proposed for ordered constraint modeling, which admits a tractable joint distribution prior. Furthermore, we are able to control the density of content vectors that violate the poset constraint by a simple "three-sigma rule". We explicitly disentangle the cross-domain images into a shared ordinal prior induced ordinal content space and two separate source/target ordinalunrelated spaces, and the self-training is worked on the shared space exclusively for ordinal-aware domain alignment. Extensive experiments on UDA medical diagnoses and facial age estimation demonstrate its effectiveness.

show abstract

Section: Related Workmentioning

confidence: 99%

Recursively Conditional Gaussian for Ordinal Unsupervised Domain Adaptation

Liu

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…More specifically, in addition to the unlabeled data instances, they have some pairs {(x (a) , x (b) )} such that for a particular factor of interest (denoted by f ), their factor values are ordered f (x (a) ) > f (x (b) ). In [23], they deal with an ordinal label problem setup while the idea is to introduce a variational posterior for the ordinal label, which is modeled as an ordinal regressor. The consequence is that, unlike our approach, they do not explicitly enforce the layout of the latent vectors to be aligned with the ordinal constraints.…”

Section: Related Workmentioning

confidence: 99%

Ordinal-Content VAE: Isolating Ordinal-Valued Content Factors in Deep Latent Variable Models

Kim¹,

Pavlović²

2020

Preprint

View full text Add to dashboard Cite

In deep representational learning, it is often desired to isolate a particular factor (termed content) from other factors (referred to as style). What constitutes the content is typically specified by users through explicit labels in the data, while all unlabeled/unknown factors are regarded as style. Recently, it has been shown that such content-labeled data can be effectively exploited by modifying the deep latent factor models (e.g., VAE) such that the style and content are well separated in the latent representations. However, the approach assumes that the content factor is categoricalvalued (e.g., subject ID in face image data, or digit class in the MNIST dataset). In certain situations, the content is ordinal-valued, that is, the values the content factor takes are ordered rather than categorical, making content-labeled VAEs, including the latent space they infer, suboptimal. In this paper, we propose a novel extension of VAE that imposes a partially ordered set (poset) structure in the content latent space, while simultaneously making it aligned with the ordinal content values. To this end, instead of the iid Gaussian latent prior adopted in prior approaches, we introduce a conditional Gaussian spacing prior model. This model admits a tractable joint Gaussian prior, but also effectively places negligible density values on the content latent configurations that violate the poset constraint. To evaluate this model, we consider two specific ordinal structured problems: estimating a subject's age in a face image and elucidating the calorie amount in a food meal image. We demonstrate significant improvements in content-style separation over previous non-ordinal approaches.

show abstract

A Novel Variational Autoencoder with Applications to Generative Modelling, Classification, and Ordinal Regression

Cited by 2 publications

References 12 publications

Recursively Conditional Gaussian for Ordinal Unsupervised Domain Adaptation

Recursively Conditional Gaussian for Ordinal Unsupervised Domain Adaptation

Ordinal-Content VAE: Isolating Ordinal-Valued Content Factors in Deep Latent Variable Models

Contact Info

Product

Resources

About