For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Multi-Task Convolutional Neural Network Leading to High Performance and Interpretability via Attribute Estimation
Keisuke MAEDA Kazaha HORII Takahiro OGAWA Miki HASEYAMA
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2020/12/01
Online ISSN: 1745-1337
Type of Manuscript: Special Section LETTER (Special Section on Smart Multimedia & Communication Systems)
Category: Neural Networks and Bioengineering
multi-task convolutional neural network, image classification, attribute estimation, interpretability,
Full Text: PDF(930.7KB)>>
A multi-task convolutional neural network leading to high performance and interpretability via attribute estimation is presented in this letter. Our method can provide interpretation of the classification results of CNNs by outputting attributes that explain elements of objects as a judgement reason of CNNs in the middle layer. Furthermore, the proposed network uses the estimated attributes for the following prediction of classes. Consequently, construction of a novel multi-task CNN with improvements in both of the interpretability and classification performance is realized.