dc.contributor.advisor | Huan, Jun | |
dc.contributor.author | Li, Xiaoli | |
dc.date.accessioned | 2019-01-01T21:06:55Z | |
dc.date.available | 2019-01-01T21:06:55Z | |
dc.date.issued | 2018-05-31 | |
dc.date.submitted | 2018 | |
dc.identifier.other | http://dissertations.umi.com/ku:15772 | |
dc.identifier.uri | http://hdl.handle.net/1808/27594 | |
dc.description.abstract | Aiming to achieve the learning capabilities possessed by intelligent beings, especially human, researchers in machine learning field have the long-standing tradition of bor- rowing ideas from human learning, such as reinforcement learning, active learning, and curriculum learning. Motivated by a philosophical theory called "constructivism", in this work, we propose a new machine learning paradigm, constructivism learning. The constructivism theory has had wide-ranging impact on various human learning theories about how human acquire knowledge. To adapt this human learning theory to the context of machine learning, we first studied how to improve leaning perfor- mance by exploring inductive bias or prior knowledge from multiple learning tasks with multiple data sources, that is multi-task multi-view learning, both in offline and lifelong setting. Then we formalized a Bayesian nonparametric approach using se- quential Dirichlet Process Mixture Models to support constructivism learning. To fur- ther exploit constructivism learning, we also developed a constructivism deep learning method utilizing Uniform Process Mixture Models. | |
dc.format.extent | 166 pages | |
dc.language.iso | en | |
dc.publisher | University of Kansas | |
dc.rights | Copyright held by the author. | |
dc.subject | Computer science | |
dc.subject | Bayesian Nonparametrics | |
dc.subject | Constructivism Learning | |
dc.subject | Multi-task Learning | |
dc.subject | Transparent Machine Learning | |
dc.title | Constructivism Learning: A Learning Paradigm for Transparent Predictive Analytics | |
dc.type | Dissertation | |
dc.contributor.cmtemember | Frost, Victor S | |
dc.contributor.cmtemember | Luo, Bo | |
dc.contributor.cmtemember | Wang, Guanghui | |
dc.contributor.cmtemember | Ho, Alfred Tat-Kei | |
dc.thesis.degreeDiscipline | Electrical Engineering & Computer Science | |
dc.thesis.degreeLevel | Ph.D. | |
dc.identifier.orcid | | |
dc.rights.accessrights | openAccess | |