Semantic uncertainty intervals for disentangled latent spaces
Meaningful uncertainty quantification in computer vision requires reasoning about semantic information -- say, the hair color of the person in a photo or the location of a car on the street. To this end, recent breakthroughs in generative modeling allow us to represent semantic information in disentangled latent spaces, but providing uncertainties on the semantic latent variables has remained challenging. In this work, we provide principled uncertainty intervals that are guaranteed to contain the true semantic factors for any underlying generative model. The method does the following: (1) it uses quantile regression to output a heuristic uncertainty interval for each element in the latent space (2) calibrates these uncertainties such that they contain the true value of the latent for a new, unseen input. The endpoints of these calibrated intervals can then be propagated through the generator to produce interpretable uncertainty visualizations for each semantic factor. This technique reliably communicates semantically meaningful, principled, and instance-adaptive uncertainty in inverse problems like image super-resolution and image completion.
Code (1)
Tasks
Image Super-Resolutionquantile regressionSuper-ResolutionUncertainty QuantificationSimilar Papers 제목 키워드 기반
Generating Sentences from Disentangled Syntactic and Semantic Spaces
Variational auto-encoders (VAEs) are widely used in natural language generation due to the regularization of the latent space. However, generating sentences from the continuous latent space does not explicitly model the …
Paraphrase GenerationText GenerationLearning Disentangled Semantic Spaces of Explanations via Invertible Neural Networks
Disentangled latent spaces usually have better semantic separability and geometrical properties, which leads to better interpretability and more controllable data generation. While this has been well investigated in Comp…
DisentanglementSentenceStyle TransferGeometry of Deep Generative Models for Disentangled Representations
Deep generative models like variational autoencoders approximate the intrinsic geometry of high dimensional data manifolds by learning low-dimensional latent-space variables and an embedding function. The geometric prope…
Representation LearningAURORA: Contextual Orthogonalization for Geometric Representation Learning in Healthcare Foundation Models
Recent healthcare foundation models have achieved strong predictive performance through large scale self supervised learning, yet their latent representations frequently entangle physiologic severity, intervention intens…
Representation LearningBoosting Latent Diffusion Models via Disentangled Representation Alignment
Latent Diffusion Models (LDMs) rely heavily on the compressed latent space provided by Variational Autoencoders (VAEs) for high-quality image generation. Recent studies have attempted to obtain generation-friendly VAEs b…
Image Generation