.png){kind=icon}
返回
Explainable machine learning in image classification models: An uncertainty quantification perspective
.png){kind=icon}
PRE
AI
DOI：10.1016/j.knosys.2022.108418.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
The poor explainability of deep learning models has hindered their adoption in safety and quality critical applications. This paper focuses on image classification models and aims to enhance the explainability of deep learning models through the development of an uncertainty quantification based framework. The proposed methodology consists of three major steps. In the first step, we adopt dropout-based Bayesian neural network to characterize the structure and parameter uncertainty inherent in deep learning models, propagate and represent such uncertainties to the model prediction as a distribution. Next, we employ entropy as a quantitative indicator to measure the uncertainty in model prediction, and develop an Empirical Cumulative Distribution Function (ECDF)-based approach to determine an appropriate threshold value for the purpose of deciding when to accept or reject the model prediction. Secondly, in the cases with high model prediction uncertainty, we combine the prediction difference analysis (PDA) approach with dropout-based Bayesian neural network to quantify the uncertainty in pixel-wise feature importance, and identify the locations in the input image that highly correlate with the model prediction uncertainty. In the third step, we develop a robustness-based design optimization formulation to enhance the relevance between input features and model prediction, and leverage a differential evolution approach to optimize the pixels in the input image with high uncertainty in feature importance. Experimental studies in MNIST and CIFAR10 image classifications are included to demonstrate the effectiveness of the proposed approach in increasing the explainability of deep learning models. (c) 2022 Elsevier B.V. All rights reserved.
Keyword:
Deep learning
Model explainability
Uncertainty quantification
Bayesian neural network
Prediction difference analysis
期刊
K
IF:
7.6
论文数:
1.3W
被引数:
4.5W
机构
引用论文
.png){kind=icon}
分享收藏分享收藏分享收藏A deep translation (GAN) based change detection network for optical and SAR remote sensing images
OAAI
OAAI分享收藏Transparency and accountability in AI decision support: Explaining and visualizing convolutional neural networks for text information人工智能决策支持中的透明度和问责制: 解释和可视化文本信息的卷积神经网络
PREAI
分享收藏Image classification with deep learning in the presence of noisy labels: A survey在存在噪声标签的情况下使用深度学习进行图像分类: 一项调查
OAAI
OAAI分享收藏
查看更多内容