Transfer learning is becoming a popular machine learning approach which aims to transfer knowledge from a source domain to a target domain. Domain adaptation is a special case of transfer learning. Domain adaptation has proven to be significant in classification tasks where the target domain does not have labelled data, a requirement for building classifiers, however, there exists a related labelled data, called as source domain.
Domain adaptation has been studied in the recent time and hence there exists many variants of domain adaptation approaches. The selection of a suitable domain adaptation approach for any given task in a real-time scenario is inevitable to achieve the desired results.
In this project, we seek to perform a comparative evaluation of various domain adaptation approaches to:
- find the main challenges in applying domain adaptation approaches
- select the most suitable approach for a given task.
You can expect to:
- learn about a range of transfer learning methods, especially domain adaptation approaches
- engage with relevant industry partners about the issues we are combatting
- communicate your work in an interactive manner
- meet regularly with your supervisor(s) to discuss ideas and research direction, as well as to receive feedback.
Skills and experience
To be considered for this project, you'll need an interest in machine learning, transfer learning and deep learning methods. Experience in machine learning and coding with Python would be beneficial.
You may be eligible to apply for a research scholarship.
Contact the supervisor for more information.