U-TraCE: a conformal prediction approach to uncertainty quantification in black-box models

Modern machine learning (ML) systems, particularly deep neural networks, have become increasingly complex and opaque, making reliable uncertainty quantification (UQ) essential. Conventional UQ methods often prove inadequate for black-box model evaluation scenarios, either because model internals are...

全面介紹

書目詳細資料

Principais autores:	Edgardo J Marchi, Maik Liebl
格式:	Artigo
語言:	Inglês
出版:	IOP Publishing 2026-01-01
叢編:	Machine Learning: Science and Technology
主題:	uncertainty quantification conformal prediction U-TraCE
在線閱讀:	https://doi.org/10.1088/2632-2153/ae35ce
標簽:	添加標簽 沒有標簽, 成為第一個標記此記錄!