arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.14430cs.LG
收藏

评估机器学习可解释性在勘探风险中的鲁棒性统一框架

A unified framework for evaluating the robustness of machine-learning interpretability for prospect risking

Center for Energy and Geo Processing (CeGP)(能源与地球处理中心);Omni Lab for Intelligent Visual Engineering and Science (OLIVES)(智能视觉工程与科学实验室);School of Electrical and Computer Engineering(电气与计算机工程学院);Georgia Institute of Technology(佐治亚理工学院)

Prithwijit Chowdhury, Ahmad Mustafa, Mohit Prabhushankar, Ghassan AlRegib

AI总结:

本文提出统一框架,通过量化必要性和充分性来评估LIME和SHAP在高维结构化勘探风险数据中的鲁棒性。

摘要:

在地球物理中,油气勘探风险涉及通过整合各种数据源的信息来评估与油气勘探相关的风险。最近,基于机器学习的分类器在表格数据上进行训练,已被用于更快地对这些勘探目标做出决策。由于此类模型决策过程的不透明性,导致了可解释人工智能(XAI)的出现。LIME和SHAP是这两种XAI方法的例子,它们试图通过按重要性对输入特征进行排名来生成特定决策的解释。然而,这些两种不同的解释策略生成的相同场景的解释却显示出分歧或不同,特别是对于复杂数据。这是因为不同解释策略对“重要性”和“相关性”的定义不同。因此,通过理论支持的因果思想中的必要性和充分性来确定这些排名特征,可以证明是提高相关解释策略可信度的更可靠和稳健的方法。我们提出一个统一的框架来生成反事实以及量化必要性和充分性,并利用这些来对LIME和SHAP提供的解释进行鲁棒性评估,以评估在高维结构化勘探风险数据上的解释。这种鲁棒性测试使我们能够更深入地了解模型处理错误数据的能力,以及哪种XAI模块最适合与哪种模型配合使用,以用于我们的油气指示数据。

英文摘要:

In geophysics, hydrocarbon prospect risking involves assessing the risks associated with hydrocarbon exploration by integrating data from various sources. Machine learning-based classifiers trained on tabular data have been recently used to make faster decisions on these prospects. The lack of transparency in the decision-making processes of such models has led to the emergence of explainable AI (XAI). LIME and SHAP are two such examples of these XAI methods which try to generate explanations of a particular decision by ranking the input features in terms of importance. However, explanations of the same scenario generated by these two different explanation strategies have shown to disagree or be different, particularly for complex data. This is because the definitions of "importance" and "relevance" differ for different explanation strategies. Thus, grounding these ranked features using theoretically backed causal ideas of necessity and sufficiency can prove to be a more reliable and robust way to improve the trustworthiness of the concerned explanation strategies.We propose a unified framework to generate counterfactuals as well as quantify necessity and sufficiency and use these to perform a robustness evaluation of the explanations provided by LIME and SHAP on high dimensional structured prospect risking data. This robustness test gives us deeper insights into the models capabilities to handle erronous data and which XAI module works best in pair with which model for our dataset for hydorcarbon indication.

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。

↑