الفهرس 
Novel active learning based approaches for balancing multi-objective maximization using trade-off between exploration and exploitation
ContentsOnly 14 pages are availabe for public view
 |  | from | 115 |  |  |
| | | from | 115 | | |
Abstract
In this thesis, we develop two novel approaches for optimization problems incurring exploration-exploitation trade-off. First, we propose a new comprehensive active learning framework including exploration-based, exploitation-based, and balancing methods. Second, we develop several analytical formulations for handling exploration-exploitation trade-off by explicitly incorporating an exploration term depending on the learning model uncertainty. We apply our proposed approaches to an operations research related application which is dynamic pricing with demand learning. We perform experiments on synthetic and real datasets. The experimental results show superior performance of our proposed approaches in terms of the achieved utility (exploitation) and estimated model error (exploration)