Lisa J Miller MS

PhD student in Computer Science, Machine Learning Group.

PhD Advisor: Dr. Susanne Still

University of Hawaii at Manoa, Information and Computer Sciences Department.

Graduate Teaching Assistant: Discrete Math I - ICS 141, Fall 2015.

Information Theoretic Learning

I work in information theoretic machine learning with Dr. Susanne Still. My interests are in predictive inference and evolutionary learning in robots and nature.

I am currently working on a project in self-organization. This work seeks to develop information theoretic methods that replicate information processing in simple biological life. There is a two-fold purpose: to increase the understanding of the role of information as a driver of evolution and emergence of biological life; and to develop techniques for constructing more self-sufficient, adaptable robotics. In my current project, I am making information theoretic self-organization algorithms more usable on complex, real-world data by incorporating approximation techniques developed for lossy compression of "Big Data".

Document Classification and Visualization.

I have recently completed a project with Dr. Rich Gazan and Dr. Still where we explored the connections between astrobiology research projects using information theoretic unsupervised learning techniques. The results of this work appear in our paper, Unsupervised Classification and Visualization of Unstructured Text for the Support of Interdisciplinary Collaboration (linked below) and were presented at the 2014 ACM Computer Supported Cooperative Work and Social Computing Conference, CSCW.

This project is part of the UH-NASA Astrobiology Institute-funded AIRFrame Project. The publications produced by astrobiology research teams over 3 years were classified and the outcomes visualized using a network visualization tool. Interactive visualizations from the project can be accessed at the AIRFrame Project Website