arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1908.06612cs.LGeess.IVstat.ML
收藏

深层神经网络还是皮肤科医生?

Deep neural network or dermatologist?

University of Queensland(昆士兰大学);Montreal Institute for Learning Algorithms(蒙特利尔学习算法研究所)

Kyle Young, Gareth Booth, Becks Simpson, Reuben Dutton, Sally Shrapnel

摘要:

深度学习技术已被证明在数字化皮肤镜图像中识别黑色素瘤的准确性很高。优点是这些方法不受人类语义预定义的特征的约束。缺点是很难理解模型预测的基本原理,也很难识别潜在的故障模式。这是在临床实践中采用深度学习的主要障碍。在本文中,我们问,如果两个现有的本地可解释性方法,梯度CAM和内核SHAP,可以揭示在黑色素瘤检测的背景下训练的卷积神经网络。我们的贡献是:(i)我们首先通过一个可复制的端到端学习框架来探索领域空间,该框架创建了一套30个模型,所有模型都是在公开的数据集上训练的(HAM10000),(ii)我们接下来通过一些基本的健全性检查实验来探索GradCAM和Kernel SHAP在此上下文中的可靠性(iii)最后,我们研究了从我们的套件中随机选择的模型,使用GradCAM和Kernel SHAP。我们发现,尽管高精度,模型偶尔会分配的重要性,不相关的诊断任务的功能。我们还表明,类似的精度模型将产生不同的解释,这些方法测量。这项工作代表了弥合皮肤癌分类领域模型准确性和可解释性之间差距的第一步。

英文摘要:

Deep learning techniques have proven high accuracy for identifying melanoma in digitised dermoscopic images. A strength is that these methods are not constrained by features that are pre-defined by human semantics. A down-side is that it is difficult to understand the rationale of the model predictions and to identify potential failure modes. This is a major barrier to adoption of deep learning in clinical practice. In this paper we ask if two existing local interpretability methods, Grad-CAM and Kernel SHAP, can shed light on convolutional neural networks trained in the context of melanoma detection. Our contributions are (i) we first explore the domain space via a reproducible, end-to-end learning framework that creates a suite of 30 models, all trained on a publicly available data set (HAM10000), (ii) we next explore the reliability of GradCAM and Kernel SHAP in this context via some basic sanity check experiments (iii) finally, we investigate a random selection of models from our suite using GradCAM and Kernel SHAP. We show that despite high accuracy, the models will occasionally assign importance to features that are not relevant to the diagnostic task. We also show that models of similar accuracy will produce different explanations as measured by these methods. This work represents first steps in bridging the gap between model accuracy and interpretability in the domain of skin cancer classification.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑