Sökning: WFRF:(Lindsten Fredrik 1984 ) >
Fast and Scalable S...
Fast and Scalable Score-Based Kernel Calibration Tests
-
- Glaser, Pierre (författare)
- University College London, Gatsby Computational Neuroscience Unit, London, UK
-
- Widmann, David (författare)
- Uppsala universitet, Avdelningen för systemteknik
-
- Lindsten, Fredrik, 1984- (författare)
- Linköpings universitet,Statistik och maskininlärning,Tekniska fakulteten
-
visa fler...
-
- Gretton, Arthur (författare)
- University College London, Gatsby Computational Neuroscience Unit, London, UK
-
visa färre...
-
(creator_code:org_t)
- 2023
- 2023
- Engelska.
-
Ingår i: Thirty-Ninth Conference on Uncertainty in Artificial Intelligence.
- Relaterad länk:
-
https://urn.kb.se/re...
Abstract
Ämnesord
Stäng
- We introduce the Kernel Calibration Conditional Stein Discrepancy test (KCCSD test), a non-parametric, kernel-based test for assessing the calibration of probabilistic models with well-defined scores. In contrast to previous methods, our test avoids the need for possibly expensive expectation approximations while providing control over its type-I error. We achieve these improvements by using a new family of kernels for score-based probabilities that can be estimated without probability density samples, and by using a conditional goodness-of-fit criterion for the KCCSD test’s U-statistic. The tractability of the KCCSD test widens the surface area of calibration measures to new promising use-cases, such as regularization during model training. We demonstrate the properties of our test on various synthetic settings.
Ämnesord
- NATURVETENSKAP -- Matematik -- Sannolikhetsteori och statistik (hsv//swe)
- NATURAL SCIENCES -- Mathematics -- Probability Theory and Statistics (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Sciences (hsv//eng)
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)