Abstract:
Multi-source domain adaptation requires reducing domain discrepancy between source domains and target domains and assessing domain relevance to determine how much knowledge should be transferred. Most previous approaches ignored domain discrepancies and relevance. This paper proposes Iterative Refinement based on Feature Selection and the Wasserstein distance (IRFSW) to solve semi-supervised domain adaptation with multiple sources. IRFSW explores domain discrepancies and relevance in an iterative learning procedure that refines learning performance until the algorithm stops. We use a sparse model to select features that reduce domain discrepancy and training loss in each iteration for each source and target domain. With the selected source and labeled target data, a classifier is created. We then calculate transferred weights using optimal transport over the selected features. The ensemble weights combine the learned classifiers to control knowledge transfer from source domains to the target domain. The method works in experiments.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
