Dr. John Hawkins is a data scientist and machine learning researcher affiliated with the Transitional AI Group out of UNSW. He has worked across multiple industries applying machine learning to business problems and published research in the application of machine learning to solve problems in molecular biology.
His current research interests are in NLP text processing and model interpretation, Bayesian Neural Networks and quantification of forecast accuracy. John is building a set of open source libraries for these and other machine learning tasks.
Minimum Viable Model Estimation
To what extent can we quantify in advance how good a machine learning system needs to be in order to solve a problem? In this work we are building tools to estimate the required performance levels of a model in order to meet business requirements specified in terms of the minimal ROI and the expected returns for predictions of differing levels of accuracy.
Interpretable Text in Machine Learning
Modern advances in interpretable machine learning are generally less applicable to text data sources. In this work we are exploring methods to render the contributions of text data more transparent regardless of the preprocessing or machine learning model that is used.
Automated Bayesian Neural Networks for Time Series Forecasting
Exploring the design of an automated Bayesian forecasting system to automate the development of time-series forecasting in a manner that provides reliable quantifictions of forecast error.
Collaborators: Dr. Rohitash Chandra Github