Dr Joshua Loftus

About

Joshua's research interests involve improving practices in data science and machine learning to reduce the impact of bias, particularly biases associated with social harms and scientific reproducibility. This includes developing methods and software for statistical inference after model selection, and using causality to analyse the fairness and interpretability of algorithms in machine learning and artificial intelligence. More broadly, he is interested in high-dimensional statistics and causal inference, and in teaching theory, applications, and best practices in data science using the R statistical programming language.