Are CNN Predictions Based on Reasonable Evidence?
Abstract
We propose Guided Zoom, an approach that utilizes spatial grounding to make more informed predictions. It does so by making sure the model has "the right reasons" for a prediction, being defined as reasons that are coherent with those used to make similar correct decisions at training time. The reason/evidence upon which a deep neural network makes a prediction is defined to be the spatial grounding, in the pixel space, for a specific class conditional probability in the model output. Guided Zoom question show reasonable the evidence used to make a prediction is. We show that Guided Zoom results in the refinement of a model's classification accuracy on two fine-grained classification datasets.
Cite
Text
Bargal et al. "Are CNN Predictions Based on Reasonable Evidence?." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2019.Markdown
[Bargal et al. "Are CNN Predictions Based on Reasonable Evidence?." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2019.](https://mlanthology.org/cvprw/2019/bargal2019cvprw-cnn/)BibTeX
@inproceedings{bargal2019cvprw-cnn,
title = {{Are CNN Predictions Based on Reasonable Evidence?}},
author = {Bargal, Sarah Adel and Zunino, Andrea and Petsiuk, Vitali and Zhang, Jianming and Saenko, Kate and Murino, Vittorio and Sclaroff, Stan},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
year = {2019},
pages = {67-70},
url = {https://mlanthology.org/cvprw/2019/bargal2019cvprw-cnn/}
}