Edwin V. Bonilla

Prestigious Scientia Scholarships on Modern Bayesian Inference in Implicit Probabilistic Models

Applications are now open for the 2019 UNSW Scientia PhD Scholarship Scheme on the project Modern Bayesian Inference in Implicit Probabilistic Models under the supervision of Dr. Edwin Bonilla, Prof. Robert Khon and Prof. Scott Sisson.

The UNSW Scientia PhD Scholarship Scheme is the most prestigious and generous scholarship scheme at UNSW and it aims to attract the best and brightest people into strategic research areas. Awardees receive a $50,000 scholarship package for four years, comprising a $40,000 per annum tax-free stipend and a travel and development support package of up to $10,000 per annum. International students also receive a tuition fee scholarship.

In addition to this scholarship package, scholars are provided with access to a range of development opportunities across research, teaching and learning and leadership and engagement.

The successful applicant(s) will also be affiliated with the Australian Research Council (ARC) Centre of Excellence for Mathematical and Statistical Frontiers (ACEMS) and will be able to access the resources of ACEMS.

Applicants should submit their expression of interest here by 20th July 2018 but are encouraged to do so as early as possible.

More info about UNSW Scientia PhD Scholarship Scheme

Project Description

Recent advances in generative adversarial networks have sparked tremendous excitement about the more general area of implicit probabilistic models. These models are only defined via simulations from an unknown (implied) distribution and provide a much more flexible data-modelling approach than traditional prescribed probabilistic models. However, their generality comes at the expense of extremely difficult inference challenges.

This project will develop new methods for inference in implicit probabilistic models and is expected to have a significant impact in areas such as evolutionary biology, ecology, high-energy physics, financial models and portfolio optimisation, where the combination of simulators with data-driven approaches is ubiquitous.

The ideal candidate should have completed a degree in a quantitative discipline such as Computer Science, Statistics, Physics, Electronic Engineering or related disciplines. He/she should have a strong Mathematics and Statistics background with excellent programming skills and proficiency in programming languages such as Python, Matlab, R or C++. He/She must be a highly-motivated student, proactive, curious and enthusiastic about scientific research.


Example papers of the type of research I do can be found at my publications page. The most closely related piece of work on relationships between GANs and Bayesian inference we have done can be found here.

Other recent work on Generic infernce in GP models, AutoGP and Gray-box Inference for Structured Gaussian Process Models are related to the aims of the project.

Excellent references providing some background and recent advances in implicit probabilistic models and GANs are give by Mohamed et al, Tran et al, and Nowozin et al.

Supervisory Team

The successful applicant will be supervised by Dr. Edwin Bonilla (Computer Science), Scientia Prof. Robert Kohn (Economics) and Prof. Scott Sisson (Mathematics and Statistics).


Feel free to contact me for more details.