Learning in High Dimensional Spaces: Applications, Theory, and Algorithms

The theoretical results are used to extend the existing learning algorithms. Based on the results from probabilistic classifiers, we have proposed an improved learning algorithm for HMMs which attempts to learn a maximum likelihood classifier under the minimum conditional entropy prior. A margin distribution optimization algorithm is proposed based on the results on generalization bounds and our results show that this new algorithm is better than the existing SVM and boosting algorithms.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Electrical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Ashutosh

Contributors dc:contributor

• Huang, Thomas S.
• Roth, Dan

• Artificial Intelligence

Language dc:language

eng

Identifier

(MiAaPQ)AAI3086006

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/80814