Probabilistic Contrastive Learning Recovers the Correct Aleatoric Uncertainty of Ambiguous Inputs

• URL: http://arxiv.org/abs/2302.02865v3
• Date: Wed, 17 May 2023 14:33:27 GMT
• Title: Probabilistic Contrastive Learning Recovers the Correct Aleatoric Uncertainty of Ambiguous Inputs
• Authors: Michael Kirchhof, Enkelejda Kasneci, Seong Joon Oh
• Abstract summary: Contrastively trained encoders have recently been proven to invert the data-generating process. We extend the common InfoNCE objective and encoders to predict latent distributions instead of points.
• Score: 21.38099300190815
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Contrastively trained encoders have recently been proven to invert the data-generating process: they encode each input, e.g., an image, into the true latent vector that generated the image (Zimmermann et al., 2021). However, real-world observations often have inherent ambiguities. For instance, images may be blurred or only show a 2D view of a 3D object, so multiple latents could have generated them. This makes the true posterior for the latent vector probabilistic with heteroscedastic uncertainty. In this setup, we extend the common InfoNCE objective and encoders to predict latent distributions instead of points. We prove that these distributions recover the correct posteriors of the data-generating process, including its level of aleatoric uncertainty, up to a rotation of the latent space. In addition to providing calibrated uncertainty estimates, these posteriors allow the computation of credible intervals in image retrieval. They comprise images with the same latent as a given query, subject to its uncertainty. Code is available at https://github.com/mkirchhof/Probabilistic_Contrastive_Learning

Related papers

• Probabilistic and Semantic Descriptions of Image Manifolds and Their Applications [28.554065677506966]
  It is common to say that images lie on a lower-dimensional manifold in the high-dimensional space. Images are unevenly distributed on the manifold, and our task is to devise ways to model this distribution as a probability distribution. We show how semantic interpretations are used to describe points on the manifold.
  arXiv  Detail & Related papers  (2023-07-06T09:36:45Z)
• SparseFusion: Distilling View-conditioned Diffusion for 3D Reconstruction [26.165314261806603]
  We propose SparseFusion, a sparse view 3D reconstruction approach that unifies recent advances in neural rendering and probabilistic image generation. Existing approaches typically build on neural rendering with re-projected features but fail to generate unseen regions or handle uncertainty under large viewpoint changes.
  arXiv  Detail & Related papers  (2022-12-01T18:59:55Z)
• What can we learn about a generated image corrupting its latent representation? [57.1841740328509]
  We investigate the hypothesis that we can predict image quality based on its latent representation in the GANs bottleneck. We achieve this by corrupting the latent representation with noise and generating multiple outputs.
  arXiv  Detail & Related papers  (2022-10-12T14:40:32Z)
• Semantic uncertainty intervals for disentangled latent spaces [30.254614465166245]
  We provide principled uncertainty intervals guaranteed to contain the true semantic factors for any underlying generative model. This technique reliably communicates semantically meaningful, principled, and instance-adaptive uncertainty in inverse problems like image super-resolution and image completion.
  arXiv  Detail & Related papers  (2022-07-20T17:58:10Z)
• Robustness via Uncertainty-aware Cycle Consistency [44.34422859532988]
  Unpaired image-to-image translation refers to learning inter-image-domain mapping without corresponding image pairs. Existing methods learn deterministic mappings without explicitly modelling the robustness to outliers or predictive uncertainty. We propose a novel probabilistic method based on Uncertainty-aware Generalized Adaptive Cycle Consistency (UGAC)
  arXiv  Detail & Related papers  (2021-10-24T15:33:21Z)
• PDC-Net+: Enhanced Probabilistic Dense Correspondence Network [161.76275845530964]
  Enhanced Probabilistic Dense Correspondence Network, PDC-Net+, capable of estimating accurate dense correspondences. We develop an architecture and an enhanced training strategy tailored for robust and generalizable uncertainty prediction. Our approach obtains state-of-the-art results on multiple challenging geometric matching and optical flow datasets.
  arXiv  Detail & Related papers  (2021-09-28T17:56:41Z)
• Uncertainty-Aware Camera Pose Estimation from Points and Lines [101.03675842534415]
  Perspective-n-Point-and-Line (Pn$PL) aims at fast, accurate and robust camera localizations with respect to a 3D model from 2D-3D feature coordinates.
  arXiv  Detail & Related papers  (2021-07-08T15:19:36Z)
• UVTomo-GAN: An adversarial learning based approach for unknown view X-ray tomographic reconstruction [27.661868972910742]
  Tomographic reconstruction recovers an unknown image given its projections from different angles. Here, we tackle a more challenging setting: 1) the projection angles are unknown, 2) they are drawn from an unknown probability distribution. In this set-up our goal is to recover the image and the projection angle distribution using an unsupervised adversarial learning approach.
  arXiv  Detail & Related papers  (2021-02-09T00:51:25Z)
• Learning Accurate Dense Correspondences and When to Trust Them [161.76275845530964]
  We aim to estimate a dense flow field relating two images, coupled with a robust pixel-wise confidence map. We develop a flexible probabilistic approach that jointly learns the flow prediction and its uncertainty. Our approach obtains state-of-the-art results on challenging geometric matching and optical flow datasets.
  arXiv  Detail & Related papers  (2021-01-05T18:54:11Z)
• Deep Bingham Networks: Dealing with Uncertainty and Ambiguity in Pose Estimation [74.76155168705975]
  Deep Bingham Networks (DBN) can handle pose-related uncertainties and ambiguities arising in almost all real life applications concerning 3D data. DBN extends the state of the art direct pose regression networks by (i) a multi-hypotheses prediction head which can yield different distribution modes. We propose new training strategies so as to avoid mode or posterior collapse during training and to improve numerical stability.
  arXiv  Detail & Related papers  (2020-12-20T19:20:26Z)
• Uncertainty Inspired RGB-D Saliency Detection [70.50583438784571]
  We propose the first framework to employ uncertainty for RGB-D saliency detection by learning from the data labeling process. Inspired by the saliency data labeling process, we propose a generative architecture to achieve probabilistic RGB-D saliency detection. Results on six challenging RGB-D benchmark datasets show our approach's superior performance in learning the distribution of saliency maps.
  arXiv  Detail & Related papers  (2020-09-07T13:01:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.