Visualization of the Basis for Decisions by Selecting Layers Based on Model's Predictions Using the Difference Between Two Networks

Takahiro Sannomiya, Kazuhiro Hotta

2024

Abstract

Grad-CAM and Score-CAM are methods to improve the interpretation of CNNs whose internal behaviour is opaque. These methods do not select which layer to use, but simply use the final layer to visualize the basis of the decision. However, we wondered whether this was really appropriate, and wondered whether there might be important information hidden in layers other than the final layer in making predictions. In the proposed method, layers are selected based on the prediction probability of the model, and the basis of judgment is visualized. In addition, by taking the difference between the model that has been trained slightly to increase the confidence level of the model’s output class and the model before training, the proposed method performs a process to emphasize the parts that contributed to the prediction and provides a better quality basis for judgment. Experimental results confirm that the proposed method outperforms existing methods in two evaluation metrics.

Paper Citation

in Harvard Style

Sannomiya T. and Hotta K. (2024). Visualization of the Basis for Decisions by Selecting Layers Based on Model's Predictions Using the Difference Between Two Networks. In Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM; ISBN 978-989-758-684-2, SciTePress, pages 370-377. DOI: 10.5220/0012419300003654

in Bibtex Style

@conference{icpram24,
author={Takahiro Sannomiya and Kazuhiro Hotta},
title={Visualization of the Basis for Decisions by Selecting Layers Based on Model's Predictions Using the Difference Between Two Networks},
booktitle={Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM},
year={2024},
pages={370-377},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012419300003654},
isbn={978-989-758-684-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM
TI - Visualization of the Basis for Decisions by Selecting Layers Based on Model's Predictions Using the Difference Between Two Networks
SN - 978-989-758-684-2
AU - Sannomiya T.
AU - Hotta K.
PY - 2024
SP - 370
EP - 377
DO - 10.5220/0012419300003654
PB - SciTePress