- Develop a principled approach to determine the choice of dimension transformation and the effects they may have upon a model.
- Develop one or more schemes to transfer effects from one dimension (i.e. feature importance) to another based on the transformation scheme.
- Assess the quality of these insights upon several real-world datasets
- Implement codes to carry out the above tasks
- Background and experience in basic Machine Learning (i.e. COMP3670/4670/4660/4650, STAT3040/4040) is required.
- Understanding of the fundamentals of Linear Algebra (i.e MATH1014/1115/1116/+ or equivalents) is required.
- Experience with Python/R/Matlab is strongly desirable
- Transferring information between dimensions: doi.org/10.1088/2632-2153/ac0167
- Generalising dimension reduction/matrix factorisiation: doi.org/10.48550/arXiv.1410.0342
- Explaning the effects that dimension reduction may have: doi.org/10.1016/j.eswa.2021.115020
- Experience in Dimension Reduction/Expansion - a vital step in many Data Analysis workflows
- Possible publications in an unexplored domain
- Coding Experience with Machine Learning Libraries
Dimension Reduction/Augmentation, Machine Learning, Data Science/Analytics