Skip to content
Enrique Valero

Enrique Valero

PhD Student

LinkedInGoogle ScholarOrcID

Enrique Valero-Leal first studied in the University of Murcia, where he got his B.Sc. in Computer Science and he then moved to Universidad Politécnica de Madrid, where he currently pursues a PhD in Artificial Intelligence
Since then, he has been participating in miscellaneous research projects, both from public funding (BAYESTREAMS) and private funding (BBVA foundation and Repsol), although he spends most of his time with his own project “Bayesian networks for interpretable machine learning”, funded by the ministry of education through the competitive fellowship for university professors training.
In addition, he has great interest in mobility and exchange program, having spent 2022 summer researching in CrossLab (Tokyo Institute of Technology) through the summer exchange research program. He is part of the ELLIS PhD program, and thus he is expected to do a long visit research visit abroad, specifically to the Department of Statistics of the Ludwig-Maximillians University of Munich.
His diverse research experience and his active participation in different projects resulted in over 5 conference/workshop articles and 1 high impact journal publication, having published alongside 7 different co-authors from 4 different institutions.

  • Research Interests:
    • His major research interests are explainable artificial intelligence and interpretable machine learning. As date of today, he focuses on how the particularities of these techniques applied to probabilistic graphical models. He has also work in the past in the fields of knowledge tracing and subgroup discovery. His primary applications are in the fields of healthcare and industry 4.0.
  • Publications:
    • Indexed Journal Articles
      • Valero-Leal, E., Bielza, C., Larrañaga, P., & Renooij, S. (2023). Efficient search for relevance explanations using MAP-independence in Bayesian networks. International Journal of Approximate Reasoning, 108965.
    • Conference Proceedings
      • Valero-Leal, E., Carlon, M. K. J., & Cross, J. S. (2023, June). A SHAP-inspired method for computing interaction contribution in deep knowledge tracing. In International Conference on Artificial Intelligence in Education (pp. 460-465). Cham: Springer Nature Switzerland.
      • Valero-Leal, E., Larrañaga, P., & Bielza, C. (2022, September). Interpreting time-varying dynamic Bayesian networks for Earth climate modelling. In International Conference on Probabilistic Graphical Models (pp. 373-384). PMLR.
    • Workshops Proceedings
      • Valero-Leal, E., Larrañaga, P., & Bielza, C. (2023). ProbExplainer: A library for unified explainability and an application in interneuron classification. In Proceedings of the Third International Workshop on Explainable AI in Healthcare.
      • Valero-Leal, E., Campos, M., & Juárez, J. M. (2022). Simple explanations to summarise subgroup discovery outcomes: A Case Study Concerning Patient Phenotyping. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases (pp. 434-451). Cham: Springer Nature Switzerland.
      • Valero-Leal, E., Larrañaga, P., & Bielza, C. (2022). Extending MAP-independence for Bayesian network explainability. In Proceedings of the First International Workshop on Heterodox Methods for Interpretable and Efficient AI.
    • Sc. Thesis
      • Valero Leal, E., Bielza, C., Larrañaga, P. (2022). Explanations for Dynamic Bayesian Networks: A Case Study in Climate Science. MSc thesis. E.T.S. de Ingenieros Informáticos, Universidad Politécnica de Madrid, 2022.
    • Sc. Thesis
      • Valero Leal, E., Campos, M., Juárez, JM. (2020). Subgroup Discovery Algorithms and Model-Agnostic Explainability. BSc thesis. Facultad de Informática, Universidad de Murcia.