.png){kind=icon}
返回
Explainable image classification with evidence counterfactual
.png){kind=icon}
OA
AI
DOI：10.1007/s10044-021-01055-y.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
The complexity of state-of-the-art modeling techniques for image classification impedes the ability to explain model predictions in an interpretable way. A counterfactual explanation highlights the parts of an image which, when removed, would change the predicted class. Both legal scholars and data scientists are increasingly turning to counterfactual explanations as these provide a high degree of human interpretability, reveal what minimal information needs to be changed in order to come to a different prediction and do not require the prediction model to be disclosed. Our literature review shows that existing counterfactual methods for image classification have strong requirements regarding access to the training data and the model internals, which often are unrealistic. Therefore, SEDC is introduced as a model-agnostic instance-level explanation method for image classification that does not need access to the training data. As image classification tasks are typically multiclass problems, an additional contribution is the introduction of the SEDC-T method that allows specifying a target counterfactual class. These methods are experimentally tested on ImageNet data, and with concrete examples, we illustrate how the resulting explanations can give insights in model decisions. Moreover, SEDC is benchmarked against existing model-agnostic explanation methods, demonstrating stability of results, computational efficiency and the counterfactual nature of the explanations.
Keyword:
Image classification
Counterfactual explanation
Explainable artificial intelligence
Search algorithms
期刊
IF:
2
论文数:
1.9K
被引数:
1.9K
机构
引用论文
Interclass testing of object oriented software.png)
Eighth IEEE International Conference on Engineering of Complex Computer Systems, 2002. Proceedings.
OAAI
Eighth IEEE International Conference on Engineering of Complex Computer Systems, 2002. Proceedings.IF0
OAAI分享收藏Serpentinized peridotites and gabbros in the Mid-Atlantic Ridge axial valley at 15°37′N and 16°52′N
PREAI
分享收藏Petrogenetic conditions at 18°–20° N MAR: Interaction between hydrothermal and magmatic systemsPetrology
PetrologyIF0
PREAI
分享收藏Unmasking Clever Hans predictors and assessing what machines really learn揭开聪明的汉斯预测器并评估机器真正学到的东西NATURE COMMUNICATIONSOAAI
NATURE COMMUNICATIONSIF15.7
OAAI分享收藏The power of obfuscation techniques in malicious JavaScript code: A measurement study恶意JavaScript代码中混淆技术的功能: 一项测量研究OAAI
OAAI分享收藏
查看更多内容