Ústav teorie informace a automatizace

Jste zde

Bibliografie

Conference Paper (international conference)

Bayesian transfer learning between Gaussian process regression tasks

Papež Milan, Quinn Anthony

: Proceedings of the IEEE International Symposium on Signal Processing and Information Technology 2019 (ISSPIT 2019)

: IEEE International Symposium on Signal Processing and Information Technology 2019 (ISSPIT 2019) /19./, (Ajman, AE, 20191209)

: GA18-15970S, GA ČR

: Bayesian transfer learning, supervised learning, fully probabilistic design, incomplete modelling, Gaussian process regression

: 10.1109/ISSPIT47144.2019.9001885

: http://library.utia.cas.cz/separaty/2019/AS/papez-0517961.pdf

(eng): Bayesian knowledge transfer in supervised learning scenarios often relies on a complete specification and optimization of the stochastic dependence between source and target tasks. This is a critical requirement of completely modelled settings, which can often be difficult to justify. We propose a strategy to overcome this. The methodology relies on fully probabilistic design to develop a target algorithm which accepts source knowledge in the form of a probability distribution. We present this incompletely modelled setting in the supervised learning context where the source and target tasks are to perform Gaussian process regression. Experimental evaluation demonstrates that the transfer of the source distribution substantially improves prediction performance of the target learner when recovering a distorted nonparametric function realization from noisy data.

: BB

: 10201

07.01.2019 - 08:39