ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Research Fellow in Machine Learning at the Australian National University

Country/Region : Australia

Website : http://internaljobs.anu.edu.au/cw/en/job...

Description

The ANU College of Engineering and Computer Science (CECS) is one of the premier engineering and computer science research institutions in the world. Comprising the Research School of Computer Science and the Research School of Engineering, both are recognised as research leaders in their respective areas continuing the tradition of excellence in research and research-led education. CECS collaborates on research projects with government departments (federal and state) and multinational corporations, helping to accelerate research and development programs and to introduce new technologies to the marketplace. CECS offers a wide range of coursework programs at both the undergraduate and graduate level, and also supervises students for higher degrees by research. We offer undergraduate degrees in engineering, information technology and computer science along with masters and doctoral postgraduate programs.
The Research School of Computer Science is seeking a Research Fellow for a fixed term period of 3 years. The appointee will work jointly with Professor Robert Williamson on an externally funded ARC research project “Uncertainty, Risk and Related Concepts in Machine Learning”. The appointee will be responsible to contribute to the aim of the project, which is to relate a wide range of theoretical concepts to each other to deepen the foundations of machine learning. These include notions of information, risk, loss, measures of uncertainty and risk, properties of distributions and aggregation functions. The proposed work is primarily theoretical. The premise is that by finding relationships between apparently different concepts, new insights can be found. The proposed work is primarily theoretical. The premise is that by finding relationships between apparently different concepts, new insights can be found. An indication of the style of work envisaged can be gleaned by looking at some of the recent papers and in particular the nature of the work is well exemplified by the older paper Information, Divergence and Risk for Binary Experiments. Details of the proposal can be found in this document (this is re-written and more up-to-date extension of the proposal for which funding has already been obtained).

Last modified: 2017-05-13 12:00:24