Sanjiv S. Jha
ABSTRACT
The increasing complexity of Cyber-Physical Systems (CPS) increases the difficulty for users to understand their behavior. Using existing Explainable Artificial Intelligence (XAI) methods, CPS can explain their behavior to the users. However, the input-output correlations used in XAI methods are not capable of explaining certain anomalies on CPS behavior caused by contextual influences (CIs) since they do not consider the context of the CPS. Some well-known techniques used for understanding such CIs on CPS are test chambers and the analysis of logged CPS data. However, test chambers are typically only available to the manufacturer of a CPS, thus not useful for understanding CIs on the shop floors. Data analysis methods focus on data correlations, which are insufficient to explain causal relationships without using expert (human) knowledge. Hence, we propose a context-aware log-based explanation system to explain the causal relationship between CIs and the behavior of a CPS. The proposed solution employs semantic technologies to access the context of the CPS. It demonstrates the causal relationship between the CPS and CIs through counterfactual explanation and abductive reasoning methods. The contextual explanations offered by the proposed system will assist users in visualizing diverse scenarios in order to improve the CPS’ behavior accordingly.
- Marcello Balduccini, Edward Griffor, Michael Huth, Claire Vishik, Martin Burns, and David Wollman. 2018. Ontology-based reasoning about the trustworthiness of cyber-physical systems. (2018).Google Scholar
- Christian Beecks, Shreekantha Devasya, and Ruben Schlutter. 2019. Machine Learning for Enhanced Waste Quantity Reduction: Insights from the MONSOON Industry 4.0 Project.Google Scholar
- Mathias Blumreiter, Joel Greenyer, Francisco Javier Chiyah Garcia, Verena Klös, Maike Schwammberger, Christoph Sommer, Andreas Vogelsang, and Andreas Wortmann. 2019. Towards self-explainable cyber-physical systems.Google Scholar
- Dieter Brughmans and David Martens. 2021. NICE: An Algorithm for Nearest Instance Counterfactual Explanations. (2021).Google Scholar
- Xiaojun Chen, Shengbin Jia, and Yang Xiang. 2020. A review: Knowledge reasoning over knowledge graph. Expert Systems with Applications 141 (2020).Google Scholar
- Andrei Ciortea, Simon Mayer, Simon Bienz, Fabien Gandon, and Olivier Corby. 2020. Autonomous search in a social and ubiquitous Web. Personal and Ubiquitous Computing(2020).Google Scholar
- Wang-Zhou Dai, Qiuling Xu, Yang Yu, and Zhi-Hua Zhou. 2019. Bridging machine learning and logical reasoning by abductive learning. (2019).Google Scholar
- Anind K Dey. 2001. Understanding and using context. Personal and ubiquitous computing 5 (2001).Google Scholar
- Lichen Fang, Yishu Yan, Ojaswi Agarwal, Shengyu Yao, Jonathan E Seppala, and Sung Hoon Kang. 2020. Effects of Environmental Temperature and Humidity on the Geometry and Strength of Polycarbonate Specimens Prepared by Fused Filament Fabrication. Materials 13(2020).Google Scholar
- Jerry Fowler and Willy Zwaenepoel. 1990. Causal distributed breakpoints. Technical Report.Google Scholar
- Andreas Holzinger, Georg Langs, Helmut Denk, Kurt Zatloukal, and Heimo Müller. 2019. Causability and explainability of artificial intelligence in medicine. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery 9 (2019).Google ScholarCross Ref
- Mark T Keane, Eoin M Kenny, Eoin Delaney, and Barry Smyth. 2021. If Only We Had Better Counterfactual Explanations: Five Key Deficits to Rectify in the Evaluation of Counterfactual XAI Techniques. (2021).Google Scholar
- Felipe Lopez, Miguel Saez, Yuru Shao, Efe C Balta, James Moyne, Z Morley Mao, Kira Barton, and Dawn Tilbury. 2017. Categorization of anomalies in smart manufacturing systems to support the selection of detection mechanisms. IEEE Robotics and Automation Letters 2 (2017).Google ScholarCross Ref
- Scott M Lundberg and Su-In Lee. 2017. A Unified Approach to Interpreting Model Predictions. In Advances in Neural Information Processing Systems 30, I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett(Eds.).Google Scholar
- Christoph Molnar. 2019. Interpretable Machine Learning.Google Scholar
- Ramaravind K Mothilal, Amit Sharma, and Chenhao Tan. 2020. Explaining machine learning classifiers through diverse counterfactual explanations. In Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency.Google ScholarDigital Library
- Nathalia Nascimento, Paulo Alencar, Carlos Lucena, and Donald Cowan. 2018. A context-aware machine learning-based approach.Google Scholar
- Leonard Petnga and Mark Austin. 2013. Ontologies of time and time-based reasoning for MBSE of cyber-physical systems. Procedia Computer Science 16 (2013).Google Scholar
- Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. ”Why Should I Trust You?”: Explaining the Predictions of Any Classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, August 13-17, 2016.Google ScholarDigital Library
- Quentin Ricard and Philippe Owezarski. 2020. Ontology Based Anomaly Detection for Cellular Vehicular Communications. In 10th European Congress on Embedded Real Time Software and Systems (ERTS 2020).Google Scholar
- Jonathan G Richens, Ciarán M Lee, and Saurabh Johri. 2020. Improving the accuracy of medical diagnosis with causal machine learning. Nature communications 11(2020).Google Scholar
- Nada Sahlab, Nasser Jazdi, and Michael Weyrich. 2020. Dynamic Context Modeling for Cyber-Physical Systems Applied to a Pill Dispenser.Google Scholar
- Tunga Salthammer. 2009. Environmental test chambers and cells. Organic indoor air pollutants: occurrence, measurement, evaluation (2009).Google ScholarCross Ref
- Udo Schlegel, Hiba Arnout, Mennatallah El-Assady, Daniela Oelke, and Daniel A Keim. 2019. Towards a rigorous evaluation of XAI Methods on Time Series. (2019).Google Scholar
- Mukesh Singhal and Ajay Kshemkalyani. 1992. An efficient implementation of vector clocks. Inform. Process. Lett. 43 (1992).Google Scholar
- Srikanth Thudumu, Philip Branch, Jiong Jin, and Jugdutt Jack Singh. 2020. A comprehensive survey of anomaly detection techniques for high dimensional big data. Journal of Big Data 7(2020).Google Scholar
- Tom Weber and Stefan Wermter. 2020. Integrating Intrinsic and Extrinsic Explainability: The Relevance of Understanding Neural Networks for Human-Robot Interaction. (2020).Google Scholar
- Adam White and Artur d’Avila Garcez. 2019. Measurable counterfactual local explanations for any classifier. (2019).Google Scholar
Index Terms
- Poster: Towards Explaining the Effects of Contextual Influences on Cyber-Physical Systems
Recommendations
Towards Independent In-Cloud Evolution of Cyber-Physical Systems
CPSNA '14: Proceedings of the 2014 IEEE International Conference on Cyber-Physical Systems, Networks, and ApplicationsThe capabilities of Cyber-Physical Systems (CPSs) are increasingly being extended towards new composite services deployed across a range of smart sensing and controlling devices. These services enable the emergence of multiple end-to-end cyber-physical ...
Towards a theory for cyber-physical systems modeling
CyPhy '14: Proceedings of the 4th ACM SIGBED International Workshop on Design, Modeling, and Evaluation of Cyber-Physical SystemsModeling the heterogeneous composition of physical, computational and communication systems is an important challenge in engineering Cyber-Physical Systems (CPS), where the major sources of heterogeneity are causality, time semantics, and different ...
Towards a Unified Framework for Cyber-Physical Systems (CPS)
CDEE '10: Proceedings of the 2010 First ACIS International Symposium on Cryptography, and Network Security, Data Mining and Knowledge Discovery, E-Commerce and Its Applications, and Embedded SystemsCyber-Physical Systems (CPS) integrate computation with physical processes. By merging computing and communication with physical processes CPS allows computer systems to monitor and interact with the physical world. However, today's computing and ...
Comments