The University of Melbourne

Statistical Machine Learning




Dr Benjamin Rubinstein investigates the robustness of the Artificial Intelligence used in today’s cyber-defences, and the impacts on user privacy of big data analysis. Ben is internationally recognised in adversarial machine learning, which seeks to answer: How can machine learning algorithms and statistical estimators be manipulated? And which algorithms are most resilient to data tampering? He has previously worked in industry research laboratories at Google, Intel, Yahoo!, Microsoft and IBM. He has also been invited to give talks at Facebook HQ and Samsung Research and has seen his research utilised in Microsoft’s Xbox game console and the Mozilla Firefox browser. His work helps to improve digital security used by citizens, industry and government, while assisting data science to balance accurate analysis with user privacy.


As well as his industry presentations, Ben’s research has been profiled on various websites, and he forms part of a committee that provides strategic policy advice to the Government and Australian organisations.