Convolutional neural networks (CNNs) for time series classification (TSC) are being increasingly used in applications ranging from quality prediction to medical diagnosis. The black box nature of these models makes understanding their prediction process difficult. This issue is crucial because CNNs are prone to learning shortcuts and biases, compromising their robustness and alignment with human expectations. To assess whether such mechanisms are being used and the associated risk, it is essential to provide model explanations that reflect the inner workings of the model. Concept Extraction (CE) methods offer such explanations, but have mostly been developed for the image domain so far, leaving a gap in the time series domain. In this work, we present a CE and localization method tailored to the time series domain, based on the ideas of CE methods for images. We propose the novel method ECLAD-ts, which provides post-hoc global explanations based on how the models encode subsets of the input at different levels of abstraction. For this, concepts are produced by clustering timestep-wise aggregations of CNN activation maps, and their importance is computed based on their impact on the prediction process. We evaluate our method on synthetic and natural datasets. Furthermore, we assess the advantages and limitations of CE in time series through empirical results. Our results show that ECLAD-ts effectively explains models by leveraging their internal representations, providing useful insights about their prediction process.
Concept Component Analysis: A Principled Approach for Concept Extraction in LLMs
Australian Institute for Machine Learning, The University of Adelaide(澳大利亚机器学习研究所,阿德莱德大学);School of Computer Science and Engineering, The University of New South Wales(计算机科学与工程学院,新南威尔士大学)
A Holistic Approach to Unifying Automatic Concept Extraction and Concept Importance Estimation
Carney Institute for Brain Science, Brown University(布朗大学卡尼脑科学研究所);Artificial and Natural Intelligence Toulouse Institute(图卢兹人工与自然智能研究所);Department of Computer Science, University of Maryland(马里兰大学计算机科学系)
Characterization and extraction of condensed representation of correlated patterns based on formal concept analysis
Université de Tunis El-Manar(突尼斯埃尔马纳尔大学);Faculté des Sciences de Tunis(突尼斯理学院);F.S.E.G de Nabeul(纳布勒经济与管理学院);Telecom Bretagne(布列塔尼电信学院);ESIG de Kairouan(凯鲁万高等信息工程学院)
Automatic Concept Extraction for Concept Bottleneck-based Video Classification
University of California Los Angeles(加州大学洛杉矶分校);University College London(伦敦大学学院);USC Information Sciences Institute(南加州大学信息科学研究所);Imperial College London(帝国理工学院);U.S. Army Research Laboratory(美国陆军研究实验室);U.S. Air Force Research Lab(美国空军研究实验室)
A Novel Biologically Mechanism-Based Visual Cognition Model--Automatic Extraction of Semantics, Formation of Integrated Concepts and Re-selection Features for Ambiguity
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Institute of Applied Mathematics, Academy of Mathematics and Systems Science, Chinese Academy of Science(中国科学院数学与系统科学研究院应用数学研究所)
Unsupervised Interpretable Basis Extraction for Concept-Based Visual Explanations
Information Technologies Institute, Centre for Research and Technology HELLAS(信息科技研究所,希腊研究中心与技术HELLOS);Department of Advanced Computing Sciences, University of Maastricht(高级计算科学系,马斯特里赫特大学)
Department of Computer Science, George Mason University(乔治梅森大学计算机科学系);Instrument Development Center, NASA Goddard Space Flight Center(美国国家航空航天局戈达德太空飞行中心仪器开发中心);Information Sciences & Technology Department, George Mason University(乔治梅森大学信息科学与技术系);George Mason University(乔治梅森大学)
Constraint-Driven Small Language Models Based on Agent and OpenAlex Knowledge Graph: Mining Conceptual Pathways and Discovering Innovation Points in Academic Papers
The Institute for Infocomm Research of A*STAR(新加坡科技研究局信息通信研究所);School of Languages and Cultures, Nanjing University of Information Science and Technology(南京信息工程大学语言文化学院)
Multi-domain Clinical Natural Language Processing with MedCAT: the Medical Concept Annotation Toolkit
King’s College London(伦敦国王学院);University College London(伦敦大学学院);Health Data Research UK(英国健康数据研究所);King’s College Hospital NHS Foundation Trust(国王学院医院国民保健服务基金会信托)
电子健康记录(EHR)包含大量非结构化文本,需要应用信息抽取(IE)技术来实现临床分析。我们提出了开源医学概念注释工具包(MedCAT),它提供:a)一种新的自监督机器学习算法,用于使用包括UMLS/SNOMED-CT在内的任何概念词汇提取概念; b)一个功能丰富的注释接口,用于定制和训练IE模型; c)集成到更广泛的CogStack生态系统中,用于与供应商无关的健康系统部署。我们在从开放数据集中提取UMLS概念方面表现出了改进的性能(F1:0.448-0.738 vs 0.429-0.650)。进一步的真实世界验证证明了伦敦3家大型医院的SNOMED-CT提取,其中自监督训练来自约1700万临床记录的约8.8B单词,并使用约6 K临床医生注释的示例进行进一步微调。我们在医院、数据集和概念类型之间显示出很强的可移植性(F1 > 0.94),这表明跨领域EHR不可知的实用性,可用于加速临床和研究用例。