El problema de la interpretabilidad de la Inteligencia Artificial y su impacto en la Administración Pública
DOI:
https://doi.org/10.36151/RCAP.3.5Keywords:
Artificial Intelligence, Public Administration, Interpretability, Explainability, TransparencyAbstract
Artificial Intelligence (AI) is one of today’s most disruptive and transformative technologies. It is increasingly present in all areas, and this massive use of technology is changing us: it is changing our education, the way we work, and the way we relate to each other and to our environment. Of course, Public Administration cannot (and should not) be oblivious to all these changes.
In the not too distant future, a large part of the decisions affecting citizens will be made by intelligent algorithms. In this scenario, the interpretability of AI algorithms will be essential to ensure their proper use in public administration.
This article aims to explain what AI interpretability is and why it is important, how this issue is being addressed from a technical point of view, what is the role of regulation in AI interpretability and what are the future prospects, especially in the field of Public Administration. In order to offer the most comprehensive approach possible, some basic fundamentals of the technology involved and the current degree of integration of AI in Europe and in our country will be explained.
Downloads
References
(Moor, 2006) Moor J. “The Dartmouth College Artificial Intelligence Conference: The Next Fifty Years” (2006). AI Magazine, 27:87-91.
(Rouse, 2024a) Rouse M. “Inteligencia artificial débil” (2024). Disponible online: https://www.techopedia.com/es/definicion/inteligencia-artificial-debil. Último acceso: 17/03/24.
(Rouse, 2024b) Rouse M. “Inteligencia artificial fuerte” (2024). Disponible online: https://www.techopedia.com/es/definicion/inteligencia-artificial-fuerte. Último acceso: 17/03/24.
(Kim et al., 2022) Kim I., Kang K., Song Y., Kim T.J. “Application of Artificial Intelligence in Pathology: Trends and Challenges” (2022). Diagnostics (Basel), 15;12(11):2794. https://doi.org/10.3390/diagnostics12112794
(Kumar et al., 2023) Kumar Y., Koul A., Singla R., Ijaz M.F. “Artificial intelligence in disease diagnosis: a systematic literature review, synthesizing framework and future research agenda” (2023). J Ambient Intell Humaniz Comput, 14(7):8459-8486. https://doi.org/10.1007/s12652-021-03612-z
(Yala et al., 2019) Yala A., Lehman C., Schuster T., Portnoi T., Barzilay R. “A deep learning mammography-based model for improved breast cancer risk prediction” (2019). Radiology, 292(1), 60-66. https://doi.org/10.1148/radiol.2019182716
(Zhou et al., 2023) Zhou X., Chen Y., Ip F.C.F. et al. “Deep learning-based polygenic risk analysis for Alzheimer’s disease prediction” (2023). Commun Med 3, 49. https://doi.org/10.1038/s43856-023-00269-x
(Gallent et al., 2023) Gallent C., Zapata A., Ortego J.L. “El impacto de la inteligencia artificial generativa en educación superior: una mirada desde la ética y la integridad académica” (2023). Relieve. 29, 2. https://doi.org/10.30827/relieve.v29i2.29134
(Franganillo, 2023) Franganillo J. “La inteligencia artificial generativa y su impacto en la creación de contenidos mediáticos” (2023). Methaodos: revista de ciencias sociales, 11 (2).
(Linardatos et al., 2020) Linardatos P., Papastefanopoulos V., Kotsiantis S. “Explainable AI: A Review of Machine Learning Interpretability Methods” (2020). Entropy (Basel), 25;23(1):18. doi: 10.3390/e23010018.
(Mariano, 2002) Mariano J.A. “Automatización de la gestión de expedientes administrativos” (2002). VII Jornadas sobre Tecnologías de la Información para la modernización de las Administraciones Públicas. Ponencia.
(Rodríguez y González, 2002) Rodríguez J.V., González J. “Integración de las tecnologías de flujo de trabajo y gestión documental para la optimización de los procesos de negocio” (2002). Ciencias de la Información, 33(3), pp. 17.
(Comisión Europea, 2019) Comisión Europea. “Prioridades de la Comisión Europea 2019-2024”. Disponible online: https://spain.representation.ec.europa.eu/estrategias-y-prioridades/prioridades-de-la-comision-europea-2019-2024_es. Último acceso: 17/03/24.
(Gobierno de España, 2024a) Gobiernto de España. “España Digital 2026”. Disponible online: https://portal.mineco.gob.es/en-us/ministerio/estrategias/Pages/00_Espana_Digital.aspx. Último acceso: 17/03/24.
(Gobierno de España, 2024b) Gobierno de España. “Catálogo de servicios de Administración digital”. Disponible online: https://administracionelectronica.gob.es/pae_Home/pae_Estrategias/Racionaliza_y_Comparte/catalogo-servicios-admon-digital.html. Último acceso: 17/03/24.
(Veale y Brass, 2019) Veale M., Brass I. “Administration by Algorithm? Public Management Meets Public Sector Machine Learning. Algorithmic Regulation” (2019). Oxford University Press.
(Restrepo-Amariles, 2020) Restrepo-Amariles D. “Algorithmic Decision Systems: Automation and Machine Learning in the Public Administration” (2020) The Cambridge Handbook of the Law of Algorithms.
(Anastasopoulos y Whitford, 2019) Anastasopoulos L.J., Whitford A.B. “Machine Learning for Public Administration Research, With Application to Organizational Reputation” (2019). Journal of Public Administration Research and Theory, 29(3), 491–510. https://doi.org/10.1093/jopart/muy060
(Henman, 2020) Henman P. “Improving public services using artificial intelligence: possibilities, pitfalls, governance” (2020). Asia Pacific Journal of Public Administration, 42:4, 209-22.
(Cerrillo i Martínez, 2019) Cerrillo i Martínez A. “El impacto de la inteligencia artificial en el derecho administrativo ¿nuevos conceptos para nuevas realidades técnicas?” (2019). Revista General de Derecho Administrativo, núm. 50.
(Etscheid, 2019) Etscheid J. “Artificial Intelligence in Public Administration” (2019). In: Electronic Government. EGOV 2019. Lecture Notes in Computer Science, vol 11685. Springer, Cham.
(Sobrino-García, 2021) Sobrino-García I. “Artificial Intelligence Risks and Challenges in the Spanish Public Administration: An Exploratory Analysis through Expert Judgements” (2021). Administrative Sciences, 1; 11(3):102.
(Cabanillas et al., 2012) Cabanillas C., Resinas M., Ruiz-Cortés A. “Automated Resource Assignment in BPMN Models Using RACI Matrices (2012). In: On the Move to Meaningful Internet Systems OTM 2012. Lecture Notes in Computer Science, vol 7565. Springer, Berlin, Heidelberg.
(Mullakara y Asokan, 2020) Mullakara N., Asokan A.K. “Robotic Process Automation Projects: Build real-world RPA solutions using UiPath and Automation Anywhere” (2020) Ed. Packt Publishing.
(Houy et al., 2019) Houy C., Hamberg M., Fettke P. “Robotic Process Automation in Public Administrations” (2019). Conference: Digitalisierung von Staat und Verwaltung. Münster, Germany.
(Uskenbayeva et al., 2019) Uskenbayeva R., Kalpeyeva Z., Satybaldiyeva R., Moldagulova A., Kassymova A. “Applying of RPA in Administrative Processes of Public Administration” (2019). IEEE 21st Conference on Business Informatics (CBI), 9-12. 10.1109/CBI.2019.10089.
(Johansson et al., 2022) Johansson J., Thomsen M., Åkesson M.A. “Public value creation and robotic process automation: normative, descriptive and prescriptive issues in municipal administration” (2022). Transforming Government: People, Process and Policy.
(Kang et al., 2020) Kang Y., Cai Z., Tan C.W., Huang Q., Liu H. “Natural language processing (NLP) in management research: A literature review” (2020). Journal of Management Analytics, 7:2, 139-172.
(Kowalski et al., 2017) Kowalski R., Esteve M., Mikhaylov S. “Application of Natural Language Processing to determine user satisfaction in Public Services” (2017). arXiv:1711.08083. https://doi.org/10.48550/arXiv.1711.08083.
(Comisión Europea, 2022) Comisión Europea. “Natural Language Processing for Public Services” (2022). Disponible online: https://joinup.ec.europa.eu/sites/default/files/inline-files/D02.01_Natural%20Language%20Processing%20for%20Public%20Services_4.pdf. Último acceso: 17/03/2024.
(Chen, 2016) Chen C.L.P. “Big Data challenges, techniques, technologies, and applications and how deep learning can be used” (2016). IEEE 20th International Conference on Computer Supported Cooperative Work in Design (CSCWD), Nanchang, China, pp. 3-3.
(Sarker, 2021) Sarker, I.H. “Deep Learning: A Comprehensive Overview on Techniques, Taxonomy, Applications and Research Directions” (2021). SN COMPUT. SCI. 2, 420. https://doi.org/10.1007/s42979-021-00815-1.
(Torres, 2020) Torres J. “Python Deep Learning. Introducción práctica con Keras y Tensorflow 2” (2020). Ed. Marcombo.
(Felderer y Ramler, 2021) Felderer M., Ramler R. “Quality Assurance for AI-Based Systems: Overview and Challenges” (2021). In: Winkler, D., Biffl, S., Mendez, D., Wimmer, M., Bergsmann, J. (eds) Software Quality: Future Perspectives on Software Engineering Quality. SWQD 2021. Lecture Notes in Business Information Processing, vol 404. Springer, Cham. https://doi.org/10.1007/978-3-030-65854-0_3.
(Martens, 2018) Martens, B. “The Importance of Data Access Regimes for Artificial Intelligence and Machine Learning“ (2018). JRC Digital Economy Working Paper 2018-09, http://dx.doi.org/10.2139/ssrn.3357652.
(Gobierno de España, 2021) Gobierno de España. “Guía al Análisis Exploratorio de Datos. Ministerio de Asuntos Económicos y Transformación Digital”. Disponible online: https://datos.gob.es/es/documentacion/guia-practica-de-introduccion-al-analisis-exploratorio-de-datos. Último acceso: 17/03/24.
(Doshi-Velez y Kim, 2017) Doshi-Velez, F., y Kim, B. (2017). Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608.
(Saeed y Omlin, 2023) Saeed W., Omlin C. “Explainable AI (XAI): A systematic meta-survey of current challenges and future opportunities” (2023) Knowledge-Based Systems, 263, 110273. https://doi.org/10.1016/j.knosys.2023.110273.
(Biran y Cotton, 2017) Biran O., Cotton C. “Explanation and justification in machine learning: A survey” (2017). In: IJCAI-17 Workshop on Explainable AI, Vol. 8, XAI, (1), 8–13.
(Gladstone AI, 2024) Gladstone AI. “An Action Plan to increase the safety and security of advanced AI” (2024). Disponible online: https://www.gladstone.ai/action-plan. Último acceso: 17/03/24.
(Woollacott, 2023) Woollacott E.” Elon Musk y otros expertos en tecnología piden parar el entrenamiento de la Inteligencia Artificial”. Forbes. Disponible online: https://forbes.es/tecnologia/256871/elon-musk-y-otros-expertos-en-tecnologia-piden-parar-el-entrenamiento-de-la-inteligencia-artificial/. Último acceso: 17/03/24.
(Han et al., 2023) Han X., Hu Z., Wang S., Zhang Y. “A Survey on Deep Learning in COVID-19 Diagnosis” (2023). J. Imaging, 9, 1. https://doi.org/10.3390/jimaging9010001
(DeGrave et al., 2021) DeGrave A.J., Janizek J.D., Lee, SI.”AI for radiographic COVID-19 detection selects shortcuts over signal” (2021). Nat Mach Intell 3, 610–619. https://doi.org/10.1038/s42256-021-00338-7
(Majeed et al, 2020) Majeed T., Rashid R., Ali D. et al. “Issues associated with deploying CNN transfer learning to detect COVID-19 from chest X-rays” (2020). Phys Eng Sci Med 43, 1289–1303. https://doi.org/10.1007/s13246-020-00934-8
(Markus et al., 2021) Markus A.F., Kors J.A., Rijnbeek P.R. “The role of explainability in creating trustworthy artificial intelligence for health care: A comprehensive survey of the terminology, design choices, and evaluation strategies” (2021). Journal of Biomedical Informatics, 113.
(Chaddad et al., 2023) Chaddad A., Peng J., Xu J., Bouridane A. “Survey of Explainable AI Techniques in Healthcare” (2023). Sensors (Basel), 5;23(2):634. doi: 10.3390/s23020634.
(Amann et al., 2020) Amann J., Blasimme A., Vayena E. et al. “Explainability for artificial intelligence in healthcare: a multidisciplinary perspective” (2020). BMC Med Inform Decis Mak 20, 310. https://doi.org/10.1186/s12911-020-01332-6.
(Algorithm Watch, 2024) Algorithm Watch. “How Dutch activists got an invasive fraud detection algorithm banned”. Disponible online: https://algorithmwatch.org/en/syri-netherlands-algorithm/. Último acceso: 17/03/24.
(Berning, 2023) Berning A.D. “El uso de sistemas basados en inteligencia artificial por las Administraciones públicas: estado actual de la cuestión y algunas propuestas ad futurum para un uso responsable” (20239. Revista de Estudios de la Administración Local y Autonómica (INAP), número 20.
(Dressel y Farid, 2018) Dressel J., Farid H. “The accuracy, fairness, and limits of predicting recidivism” (2018). Sci. Adv.4,eaao5580. DOI:10.1126/sciadv.aao5580
(Du et al., 2019) Du M., Liu N., Hu X. “Techniques for Interpretable Machine Learning” (2019). Communications of the ACM (CACM 2019). DOI:10.1145/3359786.
(Yosinski et al., 2015) Yosinski J., Clune J., Nguyen A., Fuchs T., Lipson H. “Understanding neural networks through Deep visualization (2015). In Deep Learning Workshop, ICML conference. arXiv:1506.06579. https://doi.org/10.48550/arXiv.1506.06579.
(Nguyen et al., 2016) Nguyen A., Yosinski J., Clune J. “Multifaceted feature visualization: Uncovering the different types of features learned by each neuron in deep neural networks” (2016) In Visualization for Deep Learning Workshop, ICML conference, 2016.
(Ancona et al., 2018) Ancona M., Ceolini E., Öztireli C., Gross M. “Towards better understanding of gradient-based attribution methods for Deep Neural Networks” (2018). In International Conference on Learning Representations (ICLR 2018).
(Selvaraju et al., 2017) Selvaraju R.R., Das A., Vedantam R., Cogswell M., Parikh D., Batra D. “Grad-cam: visual explanations from deep networks via gradient-based localization” (2017). International Conference on Computer Vision (ICCV 2017). https://arxiv.org/abs/1610.02391
(Wang et al., 2019) Wang H., Du M., Yang F., Zhang Z. “Score-CAM: Improved Visual Explanations Via Score-Weighted Class Activation Mapping” (2019). eprint arXiv:1910.01279. Accepted to CVPR 2020: Workshop on Fair, Data Efficient and Trusted Computer Vision. https://arxiv.org/abs/1910.01279
(Zeiler y Fergus, 2013) Zeiler M.D., Fergus R. “Visualizing and Understanding Convolutional Networks” (2013). Computer Vision and Pattern Recognition. arXiv:1311.2901. https://doi.org/10.48550/arXiv.1311.2901.
(Bach et al., 2015) Bach S., Binder A., Montavon G., Klauschen F., Müller K.R., Samek W. “On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation” (2015). PLoS ONE 10(7): e0130140. https://doi.org/10.1371/journal.pone.0130140.
(Samek et al., 2015) Samek W., Binder A., Montavon G., Bach S., Müller K.R. ·Evaluating the visualization of what a Deep Neural Network has learned· (2015). Computer Vision and Pattern Recognition. arXiv:1509.06321. https://doi.org/10.48550/arXiv.1509.06321.
(Simonyan et al., 2013) Simonyan K., Vedaldi A., Zisserman A. “Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps” (2013). Computer Vision and Pattern Recognition. arXiv:1312.6034. https://doi.org/10.48550/arXiv.1312.6034
(Mahendran y Vedaldi, 2014) Mahendran A., Vedaldi A. “Understanding Deep Image Representations by Inverting Them” (2014). arXiv:1412.0035. https://doi.org/10.48550/arXiv.1412.0035.
(Shen et al., 2024) Shen X., Song Z., Zhang Z. “AFBT GAN: enhanced explainability and diagnostic performance for cognitive decline by counterfactual generative adversarial network” (2024). arXiv preprint arXiv:2403.01758.
(Sheng-Min et al., 2021) Sheng-Min S., Tien P.J., Karnin Z. "GANMEX: One-vs-one attributions using GAN-based model explainability” (2021). International Conference on Machine Learning. PMLR.
(Comisión Europea, 2021) Comisión Europea. “Excelencia y confianza en la inteligencia artificial” (2021). Disponible online; https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/europe-fit-digital-age/excellence-and-trust-artificial-intelligence_es. Último acceso: 17/03/24.
(RTVE, 2024) RTVE. “El Parlamento Europeo ratifica la primera ley de inteligencia artificial del mundo”. Disponible online: https://www.rtve.es/noticias/20240313/parlamento-europeo-aprueba-primera-ley-inteligencia-artificial-del-mundo/16013050.shtml#:~:text=El%20Pleno%20del%20Parlamento%20Europeo,previsiblemente%20en%20el%20a%C3%B1o%202026. Último acceso: 17/03/24.
(Parlamento Europeo, 2024) Parlamento Europeo. “Ley de IA de la UE: primera normativa sobre inteligencia artificial” (2024) Disponible online: https://www.europarl.europa.eu/topics/es/article/20230601STO93804/ley-de-ia-de-la-ue-primera-normativa-sobre-inteligencia-artificial. Último acceso: 17/03/24.
(Comisión Europea, 2024) Comisión Europea. “Inteligencia artificial: preguntas y respuestas” (2024). Disponible online: https://ec.europa.eu/commission/presscorner/detail/es/QANDA_21_1683. Último acceso: 17/03/24.
(Ortiz de Zárate, 2022) Ortiz de Zárate L. “Explicabilidad (de la inteligencia artificial)” (2022). Eunomía. Revista en Cultura de la Legalidad, 22, 328-344. DOI: https://doi.org/10.20318/eunomia.2022.6819
(Faes, 2020) Faes I. “El 'big data' llega a Hacienda: un súperordenador vigilará a las multinacionales” (2020). Eleconomista.es. Disponible online: https://www.eleconomista.es/legislacion/noticias/10325315/01/20/El-big-data-llega-a-Hacienda-Un-superordenador-vigilara-a-las-multinacionales.html. Último acceso: 17/03/24.
(Administració Oberta de Catalunya, 2023) Administració Oberta de Catalunya. “Recomendador de ayudas sociales - MyGov Social” (2023), Disponible online: https://www.aoc.cat/es/projecte-innovacio/recomanador-dajuts-socials-mygov-social/. Último acceso: 17/03/24.
(Anti-Fraud Knowledge Centre, 2021) Anti-Fraud Knowledge Centre (UE). ”Sistema de alerta rápida SALER” (2021). Disponible online: https://antifraud-knowledge-centre.ec.europa.eu/library-good-practices-and-case-studies/good-practices/saler-rapid-alert-system_es. Último acceso: 17/03/24.
(Policía Nacional, 2018) Policía Nacional. “La Policía Nacional pone en funcionamiento la aplicación informática VeriPol para detectar denuncias falsas” (2018). Disponible online: https://www.policia.es/_es/comunicacion_prensa_detalle.php?ID=4433&idiomaActual=es. Último acceso: 17/03/24.
(Universitat Pompeu Fabra, 2021) Universitat Pompeu Fabra. “JULIA: Justice, Fundamental Rights and Artificial Intelligence” (2021). Disponible online: https://www.julia-project.eu/. Último acceso: 17/03/24.
(Huergo, 2023) Huergo A. “Inteligencia artificial: una aproximación jurídica no catastrofista” (2023). Revista Española de Control Externo, vol. XXV, n.º 74-75, pp. 110-129.
(Boix, 2022) Boix A. “Transparencia en la utilización de inteligencia artificial por parte de la Administración” (2022). El Cronista del Estado Social y Democrático de Derecho, núm. 100.
(Mendilibar, 2023) Mendilibar P. “Redefinición de las competencias de los empleados y empleadas públicas ante el uso de la Inteligencia Artificial por la Administración Pública. Revista Documentación Administrativa” (2023). INAP, número 10.
(Filgueiras, 2021) Filgueiras F. “Inteligencia Artificial en la administración pública: ambigüedad y elección de sistemas de IA y desafíos de gobernanza digital” (2021). Revista del CLAD Reforma y Democracia, No. 79, 5-38.
(Ponce, 2019) Ponce J. “Inteligencia artificial, Derecho administrativo y reserva de humanidad: algoritmos y procedimiento administrativo debido tecnológico”. Revista General de Derecho Administrativo (Iustel), n. 50.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Revista Canaria de Administración Pública
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
