PhD Scholarship Opportunity: Fine-Grained Explainable Classification and Recognition for Ship Identification - Naval Group Scholarship

Project title "Fine-Grained Explainable Classification and Recognition for Ship Identification."

Exciting PhD opportunity with AIML and Naval Group Pacific Pty Ltd!

This project aims to develop a cutting-edge system that leverages machine learning techniques, particularly self-supervised learning and transformers, to achieve highly accurate ship identification while providing explainable results.

Fine-grained classification requires distinguishing between objects within the same category that possess subtle differences. In this project, we will utilize pre-trained models finetuned using self-supervised learning to capture the most prominent features. This approach enables our system to adapt to scenarios with limited training instances, making it more robust and versatile. Additionally, we will leverage transformers, renowned for their ability to capture long-range dependencies and contextual information, to enhance the system's fine-grained recognition capabilities.

Ensuring explainability is a crucial aspect of our research. By integrating attention mechanisms and interpretability methods specific to transformers, we will visualize and understand the model's decision-making process, building trust and providing insights into the classification results. We acknowledge the challenge of limited annotated training data in fine-grained classification and will explore strategies that effectively utilize self-supervised learning, transfer learning, and active learning approaches to address this issue.

By combining fine-grained classification, explainability, self-supervised learning, and transformers, our project aims to push the boundaries of image recognition systems. The outcomes of this research will have broad applications, with ship identification being the primary focus of this project. Ultimately, our goal is to develop a robust and interpretable system that empowers users across various domains to harness the power of machine learning, enabling accurate and informed decision-making.

If you are passionate about advancing the field of AI and ML and have a strong background in computer vision and deep learning, this PhD opportunity is for you. Join us in shaping the future of fine-grained classification and recognition for ship identification. Apply now and be part of this ground-breaking research!

Eligibility:

Applicants must be Australian citizens or permanent residents of Australia/international students who are acceptable as candidates for a PhD/MPhil degree at the University of Adelaide.

Stipend:

The scholarship will be for 3 years and has a stipend of $38,902 (indexed annually) per annum.

It is likely to be tax exempt, subject to Taxation Office approval.  Details of any terms and/or benefits can be found in the below Conditions of Award:

4. CONDITIONS OF AWARD

4.1 The Industry Partner agrees and acknowledges that the Conditions of Award for the Scholarship:

4.1.1 are determined by the University in its absolute discretion and may be updated without notice from time to time; and

4.1.2 apply to this Agreement and the payment of any scholarship amounts to the Student.

4.2 The Conditions of Award applicable to the Scholarship under this Agreement as at the Start Date are set out in the Project Details.

Enquiries:

Contact Person: Jessica Cortazzo

School/Discipline of: The Australian Institute for Machine Learning

Email: [email protected]

Expression of interest

To be considered for this scholarship, please complete the expression of interest form .