At the 2018 International Conference on Signal Processing and Machine Learning (SPML2018) held in Shanghai, China hosted by the Singapore Chapter of the U.S.-based Association for Computing Machinery (ACM), one of the most influential academic societies in the world in the field of computer science, David Ha, Information and computer Science, Master of Science in Engineering, received the Best Presentation Award.
The awarded paper proposed a method that can precisely estimate a pattern recognition system that brings a Bayesian fault state, which is an ideal state in pattern recognition, without requiring either the division of the available learning data or repetition of learning and verification. This method exploits the properties of the class boundaries in the Bayesian fault state, which had captured little attention before, and provides a new direction in pattern recognition research.

Title of the Paper

Optimality Analysis of Boundary-Uncertainty-Based Classifier Selection Method

Presenter

David Ha (Third year student of the International Science and Technology Course, Information and Computer Science, Graduate School of Science and Engineering)

Authors

David Ha

(Third year student of the International Science and Technology Course, Information and Computer Science, Graduate School of Science and Engineering)

Hideyuki Watanabe

(Collaborative researcher, Advanced Telecommunications Research Institute International)

Yuya Tomotoshi

(Second year student of Information and Computer Science, Master's Program, Graduate School of Science and Engineering)

Emilie Delattre

(Special student, Master's Program, Graduate School of Science and Engineering, Doctor of the Université de Mons・Faculté Polytechnique)

Shigeru Katagiri

(Professor, Faculty of Science and Engineering)

* The departments, etc. are at the time of writing this paper.

Conferences Proceedings

Proceedings of 2018 International Conference on Signal Processing and Machine Learning (SPML2018)