direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Masterarbeitensvorschlag

Logistic-Heaviside Multi-Class Gaussian Process Classification

You will have to develop a Gaussian process model based on the paper  "Scalable Multi-Class Gaussian Process Classification using Expectation Propagation [1]". The idea is to change the approximation in order to make the inference much more scalable, by using the logistic link and the Polya-Gamma augmentation. You should be familiar with Gaussian Processes and Variational Inference. For more information contact Theo Galy-Fajou [2]

Masterarbeiten

2014

Bayesian Inference for a Cox-Ingersoll-Ross Model with changing Parameters and Application to Finance Data [6]

Herrmann, J.

2014 Technische Universität Berlin

Link zur Publikation [7] Download Bibtex Eintrag [8]

2013

Efficient stochastic control with Kullback Leibler costs using kernel methods [9]

Sokoloski, S.

2013 TU Berlin

Link zur Publikation [10] Download Bibtex Eintrag [11]

Meta Learning in Recommendation Systems [12]

Seiler, J.

2013 TU Berlin

Link zur Publikation [13] Download Bibtex Eintrag [14]

Efficient Optimal Control for Multi-Agent Jump Processes [15]

Schlegel, A.

2013 TU Berlin

Link zur Publikation [16] Download Bibtex Eintrag [17]

Gaussian Processes for Plume Distribution Estimation with UAVs [18]

Gosmann, J.

2013 TU Berlin

Link zur Publikation [19] Download Bibtex Eintrag [20]

2012

Probabilistic Attack on Neural Cryptography [21]

Iglesias, L. F. S.

2012 TU Berlin

Link zur Publikation [22] Download Bibtex Eintrag [23]

2010

Parameter Estimation for Stochastic Reaction Processes using Sequential Monte Carlo Methods [24]

Batz, P.

2010 HU Berlin

Link zur Publikation [25] Download Bibtex Eintrag [26]

------ Links: ------

Zusatzinformationen / Extras

Direktzugang

Schnellnavigation zur Seite über Nummerneingabe

Copyright TU Berlin 2008