NICOP - Causal feature learning

Abstract

Machine learning methods including Deep Learning for feature learning have recently been successful in many applications. Nevertheless, prediction results by Deep Learning approaches are often difficult to understand. It is difficult to see which features actually contribute its prediction performance.We aim to develop data-analytic methods to understand the mechanisms of such predictive machine learning models from the viewpoint of causality, combining the ideas of feature learning and causal structure learning. This is quite useful for users to interpret the prediction results a lot and to make better decisions.We would publish journal papers and make conference presentations to present research results. We are also planning to implement our methods and distribute the codes on the web.

Document Details

Document Type
DoD Grant Award
Publication Date
Feb 03, 2017
Source ID
N629091712034

Entities

People

  • Shohei Shimizu

Organizations

  • Office of Naval Research
  • Shiga University
  • United States Navy

Tags

Fields of Study

  • Computer science

Readers

  • Neural Network Machine Learning.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks