The Machine Learning Club is a semi-formal weekly seminar hosted by the Centre for Big Data Research in Health at the University of New South Wales (CBDRH, UNSW). We explore state-of-the-art machine learning techniques and their application to health problems to promote interest and interdisciplinary collaboration, and to support capacity development in the subject. 

The Club was established in 2018 by Dr. Oscar Perez-Concha & Ms Chrianna Bharat, and we welcome anyone with an interest in machine learning, artificial intelligence, and health data (ML/AI + Health).  

Our goal in 2022 is to foster a greater interest for developing highly scalable ML solutions applicable to long medical documents containing structure free textual data.  

We are currently looking at these topics:   

  1. Natural Language Processing (NLP) 

  1. Transformer Models [1]  

  1. GPT3 [2], BERT [3], and RoBERTa [4] 

  1. Neural Differential Equations [5] 

We meet every Friday from 2 pm to 3 pm AEST (Sydney time) 
Feel free to join by contacting our leaders below (in alphabetical order).     

Chrianna Bharat 

Chrianna is a Biostatistician and PhD Candidate at the National Drug and Alcohol Research Centre. Her research interests relate to the application of time-to-event analyses, supervised learning methods and the development and validation of predictive models on linked data. In particular, her work focuses on the application of statistical techniques to identify population subgroups to support the design and implementation of tailored opioid dependence prevention and treatment. 

Contact: c.bharat@student.unsw.edu.au 

Nicholas Kuo 

Nic joined the CBDRH as a research fellow in 2022.   
His current research lies in automating synthetic data generation for  increasing the availability of clinical data.

 University homepage: https://research.unsw.edu.au/people/mr-nic-kuo 

 Synthetic data generation: https://healthgym.ai/ 

 Contact: n.kuo@unsw.edu.au 

Jessie Liu 
Jessie was a senior data engineer before studying PhD in CBDRH, her research interest is about managing uncertainty and risk factors in machine learning-based diagnosis  

Contact: jessie.liu1@unsw.edu.au

Oscar Perez-Concha 

Oscar leads a program of research that aims to answer health and medical questions through machine learning (ML) using big data from electronic medical records (EMRs).  

He convenes and teaches the courses HDAT9500 Machine Learning I and HDAT9510 Machine Learning II that are part of the MSc Health Data Science (https://cbdrh.med.unsw.edu.au/postgraduate-coursework).  

Contact: o.perezconcha@unsw.edu.au

Juan C Quiroz 

Juan is a research fellow at CBDRH. His research interests include machine learning for diagnosis and prognosis, including applications to mental health and addiction.  

Contact: juan.quiroz@unsw.edu.au 

A special thanks to Chrianna Bharat that co-led the ML Club since its creation until May 2022.  

References:   

[1] Attention is All You Need. Vaswani et al., 2017. NeurIPS 
[2] Language Models are Few-Shot Learners. Brown et al., 2020. NeurIPS 
[3] BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding. Kenton et al., 2019. NAACL-HLT 
[4] RoBERTa: A Robustly Optimised BERT Pre-Training Approach. Liu et al., 2019. arXiv preprint arXiv:1907.11692 
[5] Neural Ordinary Differential Equations. Chen et al., 2018. NeurIPS