A compact network learning model for distribution regression.

Journal: Neural networks : the official journal of the International Neural Network Society
Published Date: Feb 1, 2019

Abstract

Despite the superior performance of deep learning in many applications, challenges remain in the area of regression on function spaces. In particular, neural networks are unable to encode function inputs compactly as each node encodes just a real value. We propose a novel idea to address this shortcoming: to encode an entire function in a single network node. To that end, we design a compact network representation that encodes and propagates functions in single nodes for the distribution regression task. Our proposed distribution regression network (DRN) achieves higher prediction accuracies while using fewer parameters than traditional neural networks.

Authors

  • Connie Khor Li Kou
    School of Computing, National University of Singapore, 13 Computing Drive, 117417, Singapore; Bioinformatics Institute, Agency for Science, Technology and Research (A*STAR), 30 Biopolis Street, 138671, Singapore. Electronic address: koukl@comp.nus.edu.sg.
  • Hwee Kuan Lee
    Daniel Aitor Holdbrook, Malay Singh, Emarene Mationg Kalaw, and Hwee Kuan Lee, Bioinformatics Institute; Malay Singh and Hwee Kuan Lee, National University of Singapore; Yukti Choudhury and Min-Han Tan, Lucence Diagnostics; Yukti Choudhury and Min-Han Tan, Institute of Bioengineering and Nanotechnology; Valerie Koh, Puay Hoon Tan, and John Yuen Shyi Peng, Singapore General Hospital; Hui Shan Tan, Ravindran Kanesvaran, and Min-Han Tan, National Cancer Center Singapore; and Hwee Kuan Lee, Institute for Infocomm Research, Singapore.
  • Teck Khim Ng
    School of Computing, National University of Singapore, 13 Computing Drive, 117417, Singapore. Electronic address: ngtk@comp.nus.edu.sg.

Keywords

External Resources

Popular Topics
Recent Journals