SearchSkip to Search Results
- 45Machine learning
- 5Artificial intelligence
- 3Online learning
- 3Reinforcement learning
- 2White, Martha
- 1Ahmed, Eisha
- 1Ajallooeian, Mohammad Mahdi
- 1Allen, Felicity R
- 1Anvik, J.
- 1Araya, Ruben
- 41Graduate Studies and Research, Faculty of
- 41Graduate Studies and Research, Faculty of/Theses and Dissertations
- 1Biological Sciences, Department of
- 1Biological Sciences, Department of/Journal Articles (Biological Sciences)
- 1Computing Science, Department of
- 1Computing Science, Department of/Technical Reports (Computing Science)
Gaussian processes are flexible probabilistic models for regression and classification. However, their success hinges on a well-specified kernel that can capture the structure of data. For complex data, the task of hand crafting a kernel becomes daunting. In this thesis, we propose new methods...
Multiple studies have reported results of activities focused on development of different algorithms for prognosis of failures of apparatuses and machines. Failure prediction allows to schedule maintenance activities, increase productivity, and decrease inventory of spare parts. Construction of...
Intrinsically disordered regions (IDRs) in proteins lack stable three dimensional structure under physiological conditions. IDRs are prevalent in nature, functionally important, and difficult to characterize experimentally due to their unstructuredness. As a result, many computational methods...
Decision-making problems with two agents can be modeled as two player games, and a Nash equilibrium is the basic solution concept describing good play in adversarial games. Computing this equilibrium solution for imperfect information games, where players have private, hidden information, is...
Cash is the most important resource at the disposal of a construction company, and management of cash has a direct impact on long- and short-term company performance. Accurately predicting the amount of cash-flow expected from particular construction operations, however, remains challenging due...
The challenge of applying machine learning techniques to diagnose schizophrenia using multi-site fMRI dataDownload
One of the main challenges for the use of machine learning techniques in neuroimaging data is the small n, large p problem. Datasets usually contain only a few hundreds of instances (n), each of which is described using hundreds of thousands of features (p). In this dissertation, we explore the...
Massive amounts of user behavior logs and sensor data are generated on mobile devices, which can help to improve the usability of social media apps and other intelligent apps. However, collecting such personal data may spark privacy and legal concerns. Recently, many efforts in both academia and...
Breaking the wall of silence of trees Mining metabolomics to describe hybridization and predict performance in the Populus – Sphaerulina musiva pathosystemDownload
Stem canker diseases in poplars caused by the fungal pathogen Sphaerulina musiva Peck. remain some of the least understood forest diseases despite causing considerable damage, particularly in hybrid poplar plantations. S. musiva is endemic to eastern Canada in provinces including the Maritimes,...
The importance of anchor ice transport of sediment on a river is significantly understudied. This study addresses the lack of data related to anchor ice release and rafting. A large sample set of anchor ice was collected in the field over the 2015-2016 and 2016-2017 winter seasons to compute a...
On the one hand, theoretical analyses of machine learning algorithms are typically performed based on various probabilistic assumptions about the data. While these probabilistic assumptions are important in the analyses, it is debatable whether such assumptions actually hold in practice. Another...