Presentation information

General Session

General Session » [GS] J-2 Machine learning

[4I2-J-2] Machine learning: uncertainty of targets

Fri. Jun 7, 2019 12:00 PM - 1:20 PM Room I (306+307 Small meeting rooms)

Chair:Kazuhiro Hotta Reviewer:Akisato Kimura

12:00 PM - 12:20 PM

[4I2-J-2-01] Importance of Uncertainty Estimation in Deep Learning

〇Iwao Maeda1, Hiroyasu Matsushima1, Hiroki Sakaji1, Kiyoshi Izumi1, David deGraw2, Hirokazu Tomioka2, Atsuo Kato3, Michiharu Kitano3 (1. The University of Tokyo, 2. Daiwa Securities Co. Ltd., 3. Daiwa Institute of Research Ltd.)

Keywords:Deep Learning, Uncertainty, Image Recognition

In recent years, predictions by machine learning and deep learning methods are utilized in various scenes of society. A model trained with deep learning methods can predict the target with high accuracy, but can not consider the predictive confidence sufficiently, and may predict high confident for extrapolated data which is hard to predict. In this study, we applied ordinary deep learning methods and methods considering predictive uncertainty, proposed in recent years, to an image classification task, and verified the robustness of trained models against extrapolated data. Models trained with the ordinary deep learning methods predicted high confidence values for data having characteristics not existing in the training data, but models trained with the methods considering uncertainty predicted low confidence values for such data. By using methods considering uncertainty, it is possible to avoid mispredictions for extrapolated data. Experimental results suggest the importance of uncertainty estimation in deep learning.