Leonardo Arrighi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Leonardo Arrighi has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Sylvio Barbon Junior
Università degli Studi di Trieste
11 shared publications - Marco Zullich
University of Groningen
3 shared publications - Michele Simonato
ASAC srl
3 shared publications - Douglas Fernandes Barbin
Department of Food Engineering, School of Food Engineering, University of Campinas, Campinas, Brazil
2 shared publications - Gabriel Marques Tavares
LMU Munich
2 shared publications - Ingrid Alves de Moraes
University of Campinas
2 shared publications
Latest Publications
- Open Access
- 17 November 2026
- Published by Springer Nature
- p. 70-89
On the Properties of Feature Attribution for Supervised Contrastive Learning
- 17 November 2026
- Published by Springer Nature
- p. 70-89
Open AccessOn the Properties of Feature Attribution for Supervised Contrastive Learning
- Open Access
- 17 November 2026
- Published by Springer Nature
- p. 277-298
How Stable Is SHAP Under Oversampling? A Study of Explanations in Imbalanced Scenarios
- 17 November 2026
- Published by Springer Nature
- p. 277-298
Open AccessHow Stable Is SHAP Under Oversampling? A Study of Explanations in Imbalanced Scenarios
- Open Access
- 17 November 2026
- Published by Springer Nature
- p. 264-284
Causal-Feature Recovery via Decision Predicate Graphs
- 17 November 2026
- Published by Springer Nature
- p. 264-284
Open AccessCausal-Feature Recovery via Decision Predicate Graphs
- Top Cited PapersOpen Access
- 24 April 2026
- Published by Springer Nature in Artificial Intelligence Review
- Vol. 59 (8)
Explainable artificial intelligence techniques for interpretation of food models: a review
- 24 April 2026
- Published by Springer Nature in Artificial Intelligence Review
- Vol. 59 (8)
Top Cited PapersOpen AccessExplainable artificial intelligence techniques for interpretation of food models: a review
- Open Access
Explaining AutoClustering: Uncovering Meta-Feature Contribution in AutoML for Clustering
AutoClustering methods aim to automate unsupervised learning tasks, including algorithm selection (AS), hyperparameter optimization (HPO), and pipeline synthesis (PS), by often leveraging meta-learning over dataset meta-features. While these systems often achieve strong performance, their recommendations are often difficult to justify: the influence of dataset meta-features on algorithm and hyperparameter choices is typically not exposed, limiting reliability, bias diagnostics, and efficient meta-feature engineering. This limits reliability and diagnostic insight for further improvements. In this work, we investigate the explainability of the meta-models in AutoClustering. We first review 22 existing methods and organize their meta-features into a structured taxonomy. We then apply a global explainability technique (i.e., Decision Predicate Graphs) to assess feature importance within meta-models from selected frameworks. Finally, we use local explainability tools such as SHAP (SHapley Additive exPlanations) to analyse specific clustering decisions. Our findings highlight consistent patterns in meta-feature relevance, identify structural weaknesses in current meta-learning strategies that can distort recommendations, and provide actionable guidance for more interpretable Automated Machine Learning (AutoML) design. This study therefore offers a practical foundation for increasing decision transparency in unsupervised learning automation.
Open AccessExplaining AutoClustering: Uncovering Meta-Feature Contribution in AutoML for Clustering
AutoClustering methods aim to automate unsupervised learning tasks, including algorithm selection (AS), hyperparameter optimization (HPO), and pipeline synthesis (PS), by often leveraging meta-learning over dataset meta-features. While these systems often achieve strong performance, their recommendations are often difficult to justify: the influence of dataset meta-features on algorithm and hyperparameter choices is typically not exposed, limiting reliability, bias diagnostics, and efficient meta-feature engineering. This limits reliability and diagnostic insight for further improvements. In this work, we investigate the explainability of the meta-models in AutoClustering. We first review 22 existing methods and organize their meta-features into a structured taxonomy. We then apply a global explainability technique (i.e., Decision Predicate Graphs) to assess feature importance within meta-models from selected frameworks. Finally, we use local explainability tools such as SHAP (SHapley Additive exPlanations) to analyse specific clustering decisions. Our findings highlight consistent patterns in meta-feature relevance, identify structural weaknesses in current meta-learning strategies that can distort recommendations, and provide actionable guidance for more interpretable Automated Machine Learning (AutoML) design. This study therefore offers a practical foundation for increasing decision transparency in unsupervised learning automation.
- 13 January 2026
- Published by Springer Nature
- p. 489-496
Discriminating Short-Term Moisture Changes in Stuffed Pasta Using Deep Computer Vision
- 13 January 2026
- Published by Springer Nature
- p. 489-496
Discriminating Short-Term Moisture Changes in Stuffed Pasta Using Deep Computer Vision
- Open Access
- 16 October 2025
- Published by Springer Nature
- p. 271-293
Extending Decision Predicate Graphs for Comprehensive Explanation of Isolation Forest
The need to explain predictive models is well-established in modern machine learning. However, beyond model interpretability, understanding pre-processing methods is equally essential. Understanding how data modifications impact model performance improvements and potential biases and promoting a reliable pipeline is mandatory for developing robust machine learning solutions. Isolation Forest (iForest) is a widely used technique for outlier detection that performs well. Its effectiveness increases with the number of tree-based learners. However, this also complicates the explanation of outlier selection and the decision boundaries for inliers. This research introduces a novel Explainable AI (XAI) method, tackling the problem of global explainability. In detail, it aims to offer a global explanation for outlier detection to address its opaque nature. Our approach is based on the Decision Predicate Graph (DPG), which clarifies the logic of ensemble methods and provides both insights and a graph-based metric to explain how samples are identified as outliers using the proposed Inlier-Outlier Propagation Score (IOP-Score). Our proposal enhances iForest’s explainability and provides a comprehensive view of the decision-making process, detailing which features contribute to outlier identification and how the model utilizes them. This method advances the state-of-the-art by providing insights into decision boundaries and a comprehensive view of holistic feature usage in outlier identification.—thus promoting a fully explainable machine learning pipeline.
- 16 October 2025
- Published by Springer Nature
- p. 271-293
Open AccessExtending Decision Predicate Graphs for Comprehensive Explanation of Isolation Forest
The need to explain predictive models is well-established in modern machine learning. However, beyond model interpretability, understanding pre-processing methods is equally essential. Understanding how data modifications impact model performance improvements and potential biases and promoting a reliable pipeline is mandatory for developing robust machine learning solutions. Isolation Forest (iForest) is a widely used technique for outlier detection that performs well. Its effectiveness increases with the number of tree-based learners. However, this also complicates the explanation of outlier selection and the decision boundaries for inliers. This research introduces a novel Explainable AI (XAI) method, tackling the problem of global explainability. In detail, it aims to offer a global explanation for outlier detection to address its opaque nature. Our approach is based on the Decision Predicate Graph (DPG), which clarifies the logic of ensemble methods and provides both insights and a graph-based metric to explain how samples are identified as outliers using the proposed Inlier-Outlier Propagation Score (IOP-Score). Our proposal enhances iForest’s explainability and provides a comprehensive view of the decision-making process, detailing which features contribute to outlier identification and how the model utilizes them. This method advances the state-of-the-art by providing insights into decision boundaries and a comprehensive view of holistic feature usage in outlier identification.—thus promoting a fully explainable machine learning pipeline.
- Open Access
Extending Decision Predicate Graphs for Comprehensive Explanation of Isolation Forest
The need to explain predictive models is well-established in modern machine learning. However, beyond model interpretability, understanding pre-processing methods is equally essential. Understanding how data modifications impact model performance improvements and potential biases and promoting a reliable pipeline is mandatory for developing robust machine learning solutions. Isolation Forest (iForest) is a widely used technique for outlier detection that performs well. Its effectiveness increases with the number of tree-based learners. However, this also complicates the explanation of outlier selection and the decision boundaries for inliers. This research introduces a novel Explainable AI (XAI) method, tackling the problem of global explainability. In detail, it aims to offer a global explanation for outlier detection to address its opaque nature. Our approach is based on the Decision Predicate Graph (DPG), which clarifies the logic of ensemble methods and provides both insights and a graph-based metric to explain how samples are identified as outliers using the proposed Inlier-Outlier Propagation Score (IOP-Score). Our proposal enhances iForest's explainability and provides a comprehensive view of the decision-making process, detailing which features contribute to outlier identification and how the model utilizes them. This method advances the state-of-the-art by providing insights into decision boundaries and a comprehensive view of holistic feature usage in outlier identification. -- thus promoting a fully explainable machine learning pipeline.
Open AccessExtending Decision Predicate Graphs for Comprehensive Explanation of Isolation Forest
The need to explain predictive models is well-established in modern machine learning. However, beyond model interpretability, understanding pre-processing methods is equally essential. Understanding how data modifications impact model performance improvements and potential biases and promoting a reliable pipeline is mandatory for developing robust machine learning solutions. Isolation Forest (iForest) is a widely used technique for outlier detection that performs well. Its effectiveness increases with the number of tree-based learners. However, this also complicates the explanation of outlier selection and the decision boundaries for inliers. This research introduces a novel Explainable AI (XAI) method, tackling the problem of global explainability. In detail, it aims to offer a global explanation for outlier detection to address its opaque nature. Our approach is based on the Decision Predicate Graph (DPG), which clarifies the logic of ensemble methods and provides both insights and a graph-based metric to explain how samples are identified as outliers using the proposed Inlier-Outlier Propagation Score (IOP-Score). Our proposal enhances iForest's explainability and provides a comprehensive view of the decision-making process, detailing which features contribute to outlier identification and how the model utilizes them. This method advances the state-of-the-art by providing insights into decision boundaries and a comprehensive view of holistic feature usage in outlier identification. -- thus promoting a fully explainable machine learning pipeline.
- Open Access
- 18 March 2025
- Published by Qeios Ltd in Qeios Preprints
Pioneering Spinal Anesthesia: A Historical Examination of Benedetto Schiassi’s Landmark Work
This article investigates the first documented use of spinal anesthesia in Italy, performed by surgeon Benedetto Schiassi in 1899. Faced with the limitations and risks of general anesthesia in a critically ill patient requiring urgent leg amputation, the surgeon employed cocaine as a local anesthetic, refining the technique with innovative modifications, including reduced dosage and prone positioning to mitigate adverse effects. Archival records and period medical publications highlight these refined techniques, contributing to the broader evolution of loco-regional anesthesia. Drawing from these historical sources, the article contextualizes Schiassi’s work within the global development of spinal anesthesia. His approach represents a significant milestone, influencing the adoption and evolution of spinal anesthesia as a safer and effective alternative to general anesthesia for surgery. By examining the broader historical and clinical significance of this event, the review highlights its impact on the trajectory of modern anesthesiology and surgical practices. This investigation offers insights into the interaction between innovation and historical medical contexts. It is intended to engage historians of medicine, anesthesiologists interested in the origins of their discipline, and surgeons involved in the evolution of anesthesiological practices useful for advances in surgical care.
- 18 March 2025
- Published by Qeios Ltd in Qeios Preprints
Open AccessPioneering Spinal Anesthesia: A Historical Examination of Benedetto Schiassi’s Landmark Work
This article investigates the first documented use of spinal anesthesia in Italy, performed by surgeon Benedetto Schiassi in 1899. Faced with the limitations and risks of general anesthesia in a critically ill patient requiring urgent leg amputation, the surgeon employed cocaine as a local anesthetic, refining the technique with innovative modifications, including reduced dosage and prone positioning to mitigate adverse effects. Archival records and period medical publications highlight these refined techniques, contributing to the broader evolution of loco-regional anesthesia. Drawing from these historical sources, the article contextualizes Schiassi’s work within the global development of spinal anesthesia. His approach represents a significant milestone, influencing the adoption and evolution of spinal anesthesia as a safer and effective alternative to general anesthesia for surgery. By examining the broader historical and clinical significance of this event, the review highlights its impact on the trajectory of modern anesthesiology and surgical practices. This investigation offers insights into the interaction between innovation and historical medical contexts. It is intended to engage historians of medicine, anesthesiologists interested in the origins of their discipline, and surgeons involved in the evolution of anesthesiological practices useful for advances in surgical care.