Greetings internet traveller. Either by luck or misfortune, you have stumbled across my home-page. People often ask me what I do. The closest approximation is mathematics. In ancient times this would have involved mucking around with triangles, circles and that sort of thing, figuring out their perimeter, area and so on. Nowadays, one mucks around with data. There are many "algorithms" whizzing around the interwebz, trying to make sense of the weird measurements computers can perform. I prove stuff about them, and sometimes suggest new ones.

I have unhinged the support vector machine, created a general framework for learning from corrupted data AND implemented it all in an open source library. Don't look up my google scholar, my ORCiD or my social security number, my papers are right here!

Born in South Africa, I have resided in Australia since 1996 and am lucky enough to call myself a citizen! I have bounced around several universities in Australia and abroad, picking up a variety of qualifications along the way.

*
Bachelor of Science (Mathematical Sciences), University of Western
Australia (2006 – 2009).
*

I specialised in the areas of pure and applied mathematics. My favourite subjects were differential geometry, algebraic topology and functional analysis.

*
Master of Science (Pure Mathematics and Statistics), University of
Melbourne (2010 – 2012).
*

It was during my masters that I began to transition from pure mathematics to statistics. I completed a thesis entitled “An Introduction to Information Geometry”.

*
PhD (Computer Science), The Australian Nation University (2012 –
2015).
*

Working under the supervision of Bob Williamson,
I completed a thesis entitled “Machine Learning via Transitions”. The
thesis outlines an abstract framework for understanding machine
learning in its *entirety*. Building on the language of statistical decision theory, the thesis
provided many novel results pertaining to learning with corrupted data.
The thesis is available
here

During my PhD I was lucky enough to present my work at the Conference on Neural Information Processing Systems (NIPS) (paper here) and at the International Conference on Machine Learning (paper here). I also spent five weeks visiting David McAllester and his team at the Toyota Technological Institute housed near the University of Chicago.

*
Research Fellowship QUT/University of California Berkeley
(2015-2017)
*

I worked with Peter Bartlett
in the broad area of machine learning theory. I also interacted with
members of the ARC Centre of Excellence for Mathematical and
Statistical Frontiers (ACEMS) and the Berkeley AI Research Group
(BAIR).

I also consulted for Dingo Software.
I helped them to scope a data centric project and to hire a data scientist.
This person is now a permanent member of staff at Dingo.

*
Research Fellowship Australian National University (September 2017
- February 2018)
*

I worked with Bob Williamson in the direction of fairness in machine learning.

I take great pride in my ability to present to technical and non-technical audiences. I have been invited to talk at several different venues. Highlights include:

- The Data Science Team at the Commonwealth Bank of Australia.
- IBM Research Australia.
- The Victorian Chapter of ANZIAM.
- A Basser Seminar at the University of Sydney.
- Lecturer AMSI Data Science Winter School.

While I am currently paid to prove theorems, I enjoy programming and have a high degree of proficiency in many languages. Though I prefer high level functional languages (F#, Scala, Haskell), I am more than happy to use whatever languages is necessary to get the job done, be it Python, Matlab, Mathematica and so on.

For proof of my programming competency, please see my open source machine learning library FLearn.