Research, teaching, projects, and technical skills.

Federation University Australia · Institute of Innovation, Science & Sustainability

• Research focus: interpretable representation learning for scientific data.
• Thesis direction: Mixture-of-Encoders framework for interpretable representation learning.
• Application domains: ophthalmic imaging and tropical meteorology.

Developing MixtureBetaVAE-style generative-discriminative models for traversable latent spaces.

• Mixture-of-encoders β-VAE architecture.
• Latent traversal and counterfactual interpretation.
• Evaluation across discrimination, reconstruction, and interpretability.

Representation learning for OCT-derived RNFL and GCIPL maps related to glaucoma progression.

Spatio-temporal tensor modelling using OWZ events and ERA5 environmental variables.

Mixture-of-encoders β-VAE framework for interpretable scientific representation learning.

Pipeline for generating model-ready cyclone seed tensors from OWZ and ERA5 data.

Workflow for preparing RNFL and GCIPL maps for glaucoma representation learning.

Add official teaching assistant, tutor, lab demonstrator, marking, or mentoring roles here.

• Machine learning labs and assignment feedback.
• Python and algorithmic problem-solving support.
• Emphasis on reproducible code and clear reasoning.