We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.ML

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: Conditional Sampling With Monotone GANs

Abstract: We present a new approach for sampling conditional probability measures, enabling consistent uncertainty quantification in supervised learning tasks. We construct a mapping that transforms a reference measure to the measure of the output conditioned on new inputs. The mapping is trained via a modification of generative adversarial networks (GANs), called monotone GANs, that imposes monotonicity and a block triangular structure. We present theoretical guarantees for the consistency of our proposed method, as well as numerical experiments demonstrating the ability of our method to accurately sample conditional measures in applications ranging from inverse problems to image in-painting.
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Computation (stat.CO)
Cite as: arXiv:2006.06755 [stat.ML]
  (or arXiv:2006.06755v2 [stat.ML] for this version)

Submission history

From: Bamdad Hosseini Dr. [view email]
[v1] Thu, 11 Jun 2020 19:15:43 GMT (1978kb,D)
[v2] Fri, 19 Feb 2021 16:36:59 GMT (3587kb,D)

Link back to: arXiv, form interface, contact.