Soon to be graduate from the Australian National University with a Bachelor of Advanced Computing (Research & Development). Specialised in Machine Learning. Currently working as the Lead Developer at Sembl.
Investigation into the use of WSSS for histopathology images to reduce the burden of manual annotation. Introduces a novel training method that has competitive performance with SOTA on the BCSS-WSSS dataset.
This thesis explores novel modifications to Cross-Batch Memory for Deep Metric Learning, addressing computational efficiency and adaptability. It introduces Per-Class, Super-Class, and Delayed Normalisation techniques, demonstrating their effectiveness in enhancing DML model performance and stability.
This paper investigates how Federated Learning (FL) can optimize scientific collaboration in materials science by simulating various data-sharing scenarios. The study concludes that sharing sample spaces is the most effective collaboration strategy for FL, outperforming scenarios where features are shared or no data is exchanged.
Download my CV to learn more about my academic background, publications, and experience.
Download CV