Knowledge graphs have become powerful sources for web search, but an equivalent source about things and their relations in pictures and videos does not exist yet. This project develops core techniques to learn image-centric knowledge graphs by connecting large collections of image/video and their descriptions to existing knowledge bases with encyclopedic, lexical, and commonsense knowledge. Such multimedia knowledge graphs will be used to advance automated understanding of text and multimedia, as well as social events and other user-generated content.
The research position seeks to advance both machine learning methods and formulating relation learning problems. The position would focus on developing novel methods and models on:
- relation learning from large scale online data.
- images to text and text to images.
- learning on graphs and sequences.
The ANU is a highly-ranked research-intensive university (world-wide rankings - QS: 25th, Times: 48th, Shanghai ARWU 66th). The postdoc research fellow will be located at the strong ANU AI research group, with frequent interactions with NICTA Machine Learning group of 20+ researchers on machine learning and data science. Career development opportunities include possible contribution to teaching, the ability to apply for independent grants with the ARC, inter-diciplinary collaborations, or work with government/industry.
The intended duration for the research fellow position is 3 years; PhD candidature at ANU CS usually last 3 to 3.5 years.
Candidates with the following background are encouraged to apply:
* a solid machine learning and mathematical background
* in-depth understanding with at least one inference paradigms
* a willingness to formulate and validate machine learning problems
* experiences working with large, real-world datasets a plus
The research will be conducted under the supervision of Lexing Xie with a number of collaborators inside and outside of the Computational Media Lab (http://cm.cecs.anu.edu.au). We accept applications until position is filled. Interested researchers may write to lexing.xie@anu.edu.au including a CV, along with a few words explaining how well their interest and technical background matches the topic areas above.