HIGHLIGHTS
- who: Franko Schmähling from the Physikal-Techn. Bundesanstalt, Abbestr., Berlin, Germany have published the research work: A framework for benchmarking uncertainty in deep regression, in the Journal: (JOURNAL)
- what: The authors propose a framework for the assessment of uncertainty quantification in deep regression. The aim of this work is to propose a systematic, and flexible, approach to the problem of creating datasets for regression problems with a known ground truth for benchmarking uncertainties in deep learning. Namely, the authors propose to use regression problems of the kind yi ~ N G(xi )T γ, σ 2, that . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.