CNR Author
operator:
and / or
Typology
operator:
and / or
Language
operator:
and / or
Date
operator:
and / or
Rights
operator:
and / or
2025
Journal article
Open Access
Explainable, effective, and efficient learning-to-rank models using ILMART
Lucchese C., Nardini F. M., Orlando S., Perego R., Veneri A.
Learning ranking models that are both explainable and effective is an emerging topic within the research area of explainable AI. Several Learning-to-Rank (LtR) algorithms have been recently proposed that build models that are simple to explain and, at the same time, almost as effective as their state-of-the-art, black-box counterparts. In this work, we propose Interpretable LambdaMART (ILMART), a novel framework with different strategies to constrain the state-of-the-art LtR LambdaMART algorithm to generate interpretable models, i.e., ensembles whose trees can use either single features (main effects) or a limited number of interacting features (interaction effects). ILMART facilitates a straightforward tradeoff between model explainability and effectiveness by precisely tuning the quantity of main and interaction effects during the learning phase. We show that slightly increasing their number allows ILMART models to reach ranking performances at par with full-complexity LambdaMART ones. Furthermore, reproducible experiments conducted on publicly available LtR datasets demonstrate that ILMART can improve nDCG@10 by up to 10% compared to state-of-the-art competitors while preserving an explainable structure. Finally, we explore the relationship between model explainability and inference efficiency by introducing a novel and easy-to-implement scoring algorithm for ILMART ranking models, achieving up to a speedup compared to the baseline.Source: ACM TRANSACTIONS ON INFORMATION SYSTEMS, vol. 43 (issue 4)
DOI: 10.1145/3733232
Project(s): EFRA
Metrics:
Lucchese C., Nardini F. M., Orlando S., Perego R., Veneri A.
Learning ranking models that are both explainable and effective is an emerging topic within the research area of explainable AI. Several Learning-to-Rank (LtR) algorithms have been recently proposed that build models that are simple to explain and, at the same time, almost as effective as their state-of-the-art, black-box counterparts. In this work, we propose Interpretable LambdaMART (ILMART), a novel framework with different strategies to constrain the state-of-the-art LtR LambdaMART algorithm to generate interpretable models, i.e., ensembles whose trees can use either single features (main effects) or a limited number of interacting features (interaction effects). ILMART facilitates a straightforward tradeoff between model explainability and effectiveness by precisely tuning the quantity of main and interaction effects during the learning phase. We show that slightly increasing their number allows ILMART models to reach ranking performances at par with full-complexity LambdaMART ones. Furthermore, reproducible experiments conducted on publicly available LtR datasets demonstrate that ILMART can improve nDCG@10 by up to 10% compared to state-of-the-art competitors while preserving an explainable structure. Finally, we explore the relationship between model explainability and inference efficiency by introducing a novel and easy-to-implement scoring algorithm for ILMART ranking models, achieving up to a speedup compared to the baseline.Source: ACM TRANSACTIONS ON INFORMATION SYSTEMS, vol. 43 (issue 4)
DOI: 10.1145/3733232
Project(s): EFRA
Metrics:
See at:
dl.acm.org 
| CNR IRIS | ACM Transactions on Information Systems 
| CNR IRIS
2024
Conference article
Open Access
On the application of a common theoretical explainability framework in information retrieval
Albarelli A., Lucchese C., Rizzo M., Veneri A.
Most of the current state-of-the-art models used to solve the search and ranking tasks in Information Retrieval (IR) are considered “black boxes” due to the enormous number of parameters employed, which makes it difficult for humans to understand the relation between input and output. Thus, in the current literature, several approaches are proposed to explain their outputs, trying to make the models more explainable while maintaining the high level of effectiveness achieved. Even though many methods have been developed, there is still a lack of a common way of describing and evaluating the models and methods of the Explainabile IR (ExIR) field. This work shows how a common theoretical framework for explainability (previously presented in the biomedical field) can be applied to IR. We first describe the general framework and then focus on specific explanation techniques in the IR field, focusing on core IR tasks: search and ranking. We show how well-known methods in ExIR fit into the framework and how specific IR explainability evaluation metrics can be described using this new setting.Source: CEUR WORKSHOP PROCEEDINGS, vol. 3802, pp. 43-52. Udine, Italy, 05-06/09/2024
Albarelli A., Lucchese C., Rizzo M., Veneri A.
Most of the current state-of-the-art models used to solve the search and ranking tasks in Information Retrieval (IR) are considered “black boxes” due to the enormous number of parameters employed, which makes it difficult for humans to understand the relation between input and output. Thus, in the current literature, several approaches are proposed to explain their outputs, trying to make the models more explainable while maintaining the high level of effectiveness achieved. Even though many methods have been developed, there is still a lack of a common way of describing and evaluating the models and methods of the Explainabile IR (ExIR) field. This work shows how a common theoretical framework for explainability (previously presented in the biomedical field) can be applied to IR. We first describe the general framework and then focus on specific explanation techniques in the IR field, focusing on core IR tasks: search and ranking. We show how well-known methods in ExIR fit into the framework and how specific IR explainability evaluation metrics can be described using this new setting.Source: CEUR WORKSHOP PROCEEDINGS, vol. 3802, pp. 43-52. Udine, Italy, 05-06/09/2024
See at:
ceur-ws.org | CNR IRIS | CNR IRIS
2024
Conference article
Open Access
Software systems compliance with the AI Act: lessons learned from an international challenge
Scantamburlo T., Falcarin P., Veneri A., Fabris A., Gallese C., Billa V., Rotolo F., Marcuzzi F.
In this experience paper, we present the lessons learned from the First University of St. Gallen Grand Challenge 2023, a competition involving interdisciplinary teams tasked with assessing the legal compliance of real-world AI-based systems with the European Union's Artificial Intelligence Act (AI Act). The AI Act is the very first attempt in the world to regulate AI systems and its potential impact is huge. The competition provided firsthand experience and practical knowledge regarding the AI Act's requirements. It also highlighted challenges and opportunities for the software engineering and AI communities.DOI: 10.1145/3643691.3648589
Metrics:
Scantamburlo T., Falcarin P., Veneri A., Fabris A., Gallese C., Billa V., Rotolo F., Marcuzzi F.
In this experience paper, we present the lessons learned from the First University of St. Gallen Grand Challenge 2023, a competition involving interdisciplinary teams tasked with assessing the legal compliance of real-world AI-based systems with the European Union's Artificial Intelligence Act (AI Act). The AI Act is the very first attempt in the world to regulate AI systems and its potential impact is huge. The competition provided firsthand experience and practical knowledge regarding the AI Act's requirements. It also highlighted challenges and opportunities for the software engineering and AI communities.DOI: 10.1145/3643691.3648589
Metrics:
See at:
dl.acm.org | CNR IRIS | Archivio istituzionale della ricerca - Università degli Studi di Venezia Ca' Foscari | doi.org | Archivio istituzionale della ricerca - Università degli Studi di Venezia Ca' Foscari | CNR IRIS
2023
Conference article
Open Access
GAM Forest explanation
Lucchese C, Orlando S, Perego R, Veneri A
Most accurate machine learning models unfortunately produce black-box predictions, for which it is impossible to grasp the internal logic that leads to a specific decision. Unfolding the logic of such black-box models is of increasing importance, especially when they are used in sensitive decision-making processes. In thisworkwe focus on forests of decision trees, which may include hundreds to thousands of decision trees to produce accurate predictions. Such complexity raises the need of developing explanations for the predictions generated by large forests.We propose a post hoc explanation method of large forests, named GAM-based Explanation of Forests (GEF), which builds a Generalized Additive Model (GAM) able to explain, both locally and globally, the impact on the predictions of a limited set of features and feature interactions.We evaluate GEF over both synthetic and real-world datasets and show that GEF can create a GAM model with high fidelity by analyzing the given forest only and without using any further information, not even the initial training dataset.DOI: 10.48786/edbt.2023.14
Metrics:
Lucchese C, Orlando S, Perego R, Veneri A
Most accurate machine learning models unfortunately produce black-box predictions, for which it is impossible to grasp the internal logic that leads to a specific decision. Unfolding the logic of such black-box models is of increasing importance, especially when they are used in sensitive decision-making processes. In thisworkwe focus on forests of decision trees, which may include hundreds to thousands of decision trees to produce accurate predictions. Such complexity raises the need of developing explanations for the predictions generated by large forests.We propose a post hoc explanation method of large forests, named GAM-based Explanation of Forests (GEF), which builds a Generalized Additive Model (GAM) able to explain, both locally and globally, the impact on the predictions of a limited set of features and feature interactions.We evaluate GEF over both synthetic and real-world datasets and show that GEF can create a GAM model with high fidelity by analyzing the given forest only and without using any further information, not even the initial training dataset.DOI: 10.48786/edbt.2023.14
Metrics:
See at:
CNR IRIS | ISTI Repository | openproceedings.org | CNR IRIS
2023
Conference article
Open Access
Can embeddings analysis explain large language model ranking?
Lucchese C, Minello G, Nardini Fm, Orlando S, Perego R, Veneri A
Understanding the behavior of deep neural networks for Information Retrieval (IR) is crucial to improve trust in these effective models. Current popular approaches to diagnose the predictions made by deep neural networks are mainly based on: i) the adherence of the retrieval model to some axiomatic property of the IR system, ii) the generation of free-text explanations, or iii) feature importance attributions. In this work, we propose a novel approach that analyzes the changes of document and query embeddings in the latent space and that might explain the inner workings of IR large pre-trained language models. In particular, we focus on predicting query/document relevance, and we characterize the predictions by analyzing the topological arrangement of the embeddings in their latent space and their evolution while passing through the layers of the network. We show that there exists a link between the embedding adjustment and the predicted score, based on how tokens cluster in the embedding space. This novel approach, grounded in the query and document tokens interplay over the latent space, provides a new perspective on neural ranker explanation and a promising strategy for improving the efficiency of the models and Query Performance Prediction (QPP).DOI: 10.1145/3583780.3615225
Metrics:
Lucchese C, Minello G, Nardini Fm, Orlando S, Perego R, Veneri A
Understanding the behavior of deep neural networks for Information Retrieval (IR) is crucial to improve trust in these effective models. Current popular approaches to diagnose the predictions made by deep neural networks are mainly based on: i) the adherence of the retrieval model to some axiomatic property of the IR system, ii) the generation of free-text explanations, or iii) feature importance attributions. In this work, we propose a novel approach that analyzes the changes of document and query embeddings in the latent space and that might explain the inner workings of IR large pre-trained language models. In particular, we focus on predicting query/document relevance, and we characterize the predictions by analyzing the topological arrangement of the embeddings in their latent space and their evolution while passing through the layers of the network. We show that there exists a link between the embedding adjustment and the predicted score, based on how tokens cluster in the embedding space. This novel approach, grounded in the query and document tokens interplay over the latent space, provides a new perspective on neural ranker explanation and a promising strategy for improving the efficiency of the models and Query Performance Prediction (QPP).DOI: 10.1145/3583780.3615225
Metrics:
See at:
CNR IRIS | ISTI Repository | CNR IRIS
2023
Conference article
Open Access
Explaining learning to rank methods to improve them
Veneri A
State-of-the-art methods for Learning to Rank (LtR), either designed for tabular or textual data, are incredibly complex. Increasing the complexity of the models has many drawbacks, including difficulties in understanding the logic behind each prediction and a lack of trust in the system during its deployment. In this paper, which describes the author's goals during his Ph.D., there is an analysis and discussion of how we can use the ideas and tools coming from the eXplainable Artificial Intelligence (XAI) field to make the most effective methods for LtR understandable to the practitioners with the final goal of making them more efficient and/or understand better when they can be improved. The strategies adopted to achieve the aforementioned goals are different and based on the type of models analyzed, which go from more traditional LtR models based on ensembles of decision trees and using handcrafted features to fairly new neural LtR models using text data.DOI: 10.1145/3583780.3616002
Metrics:
Veneri A
State-of-the-art methods for Learning to Rank (LtR), either designed for tabular or textual data, are incredibly complex. Increasing the complexity of the models has many drawbacks, including difficulties in understanding the logic behind each prediction and a lack of trust in the system during its deployment. In this paper, which describes the author's goals during his Ph.D., there is an analysis and discussion of how we can use the ideas and tools coming from the eXplainable Artificial Intelligence (XAI) field to make the most effective methods for LtR understandable to the practitioners with the final goal of making them more efficient and/or understand better when they can be improved. The strategies adopted to achieve the aforementioned goals are different and based on the type of models analyzed, which go from more traditional LtR models based on ensembles of decision trees and using handcrafted features to fairly new neural LtR models using text data.DOI: 10.1145/3583780.3616002
Metrics:
See at:
dl.acm.org | CNR IRIS | CNR IRIS
2023
Conference article
Open Access
A theoretical framework for AI models explainability with application in biomedicine
Rizzo M, Veneri A, Albarelli A, Lucchese C, Nobile M, Conati C
EXplainable Artificial Intelligence (XAI) is a vibrant research topic in the artificial intelligence community. It is raising growing interest across methods and domains, especially those involving high stake decision-making, such as the biomedical sector. Much has been written about the subject, yet XAI still lacks shared terminology and a framework capable of providing structural soundness to explanations. In our work, we address these issues by proposing a novel definition of explanation that synthesizes what can be found in the literature. We recognize that explanations are not atomic but the combination of evidence stemming from the model and its input-output mapping, and the human interpretation of this evidence. Furthermore, we fit explanations into the properties of faithfulness (i.e., the explanation is an accurate description of the model's inner workings and decision-making process) and plausibility (i.e., how much the explanation seems convincing to the user). Our theoretical framework simplifies how these properties are operationalized, and it provides new insights into common explanation methods that we analyze as case studies. We also discuss the impact that our framework could have in biomedicine, a very sensitive application domain where XAI can have a central role in generating trust.DOI: 10.1109/cibcb56990.2023.10264877
Metrics:
Rizzo M, Veneri A, Albarelli A, Lucchese C, Nobile M, Conati C
EXplainable Artificial Intelligence (XAI) is a vibrant research topic in the artificial intelligence community. It is raising growing interest across methods and domains, especially those involving high stake decision-making, such as the biomedical sector. Much has been written about the subject, yet XAI still lacks shared terminology and a framework capable of providing structural soundness to explanations. In our work, we address these issues by proposing a novel definition of explanation that synthesizes what can be found in the literature. We recognize that explanations are not atomic but the combination of evidence stemming from the model and its input-output mapping, and the human interpretation of this evidence. Furthermore, we fit explanations into the properties of faithfulness (i.e., the explanation is an accurate description of the model's inner workings and decision-making process) and plausibility (i.e., how much the explanation seems convincing to the user). Our theoretical framework simplifies how these properties are operationalized, and it provides new insights into common explanation methods that we analyze as case studies. We also discuss the impact that our framework could have in biomedicine, a very sensitive application domain where XAI can have a central role in generating trust.DOI: 10.1109/cibcb56990.2023.10264877
Metrics:
2022
Conference article
Open Access
ILMART: interpretable ranking with constrained LambdaMART
Lucchese C, Nardini Fm, Orlando S, Perego R, Veneri A
Interpretable Learning to Rank (LtR) is an emerging field within the research area of explainable AI, aiming at developing intelligible and accurate predictive models. While most of the previous research efforts focus on creating post-hoc explanations, in this paper we investigate how to train effective and intrinsically-interpretable ranking models. Developing these models is particularly challenging and it also requires finding a trade-off between ranking quality and model complexity. State-of-the-art rankers, made of either large ensembles of trees or several neural layers, exploit in fact an unlimited number of feature interactions making them black boxes. Previous approaches on intrinsically-interpretable ranking models address this issue by avoiding interactions between features thus paying a significant performance drop with respect to full-complexity models. Conversely, ILMART, our novel and interpretable LtR solution based on LambdaMART, is able to train effective and intelligible models by exploiting a limited and controlled number of pairwise feature interactions. Exhaustive and reproducible experiments conducted on three publicly-available LtR datasets show that ILMART outperforms the current state-of-the-art solution for interpretable ranking of a large margin with a gain of nDCG of up to 8%.DOI: 10.1145/3477495.3531840
Metrics:
Lucchese C, Nardini Fm, Orlando S, Perego R, Veneri A
Interpretable Learning to Rank (LtR) is an emerging field within the research area of explainable AI, aiming at developing intelligible and accurate predictive models. While most of the previous research efforts focus on creating post-hoc explanations, in this paper we investigate how to train effective and intrinsically-interpretable ranking models. Developing these models is particularly challenging and it also requires finding a trade-off between ranking quality and model complexity. State-of-the-art rankers, made of either large ensembles of trees or several neural layers, exploit in fact an unlimited number of feature interactions making them black boxes. Previous approaches on intrinsically-interpretable ranking models address this issue by avoiding interactions between features thus paying a significant performance drop with respect to full-complexity models. Conversely, ILMART, our novel and interpretable LtR solution based on LambdaMART, is able to train effective and intelligible models by exploiting a limited and controlled number of pairwise feature interactions. Exhaustive and reproducible experiments conducted on three publicly-available LtR datasets show that ILMART outperforms the current state-of-the-art solution for interpretable ranking of a large margin with a gain of nDCG of up to 8%.DOI: 10.1145/3477495.3531840
Metrics:
See at:
dl.acm.org | CNR IRIS | ISTI Repository | CNR IRIS | CNR IRIS
2022
Conference article
Open Access
Interpretable ranking using LambdaMART (Abstract)
Lucchese C, Nardini Fm, Orlando S, Perego R, Veneri A
Source: CEUR WORKSHOP PROCEEDINGS. Milano, Italy, 29-30/06/2022
Lucchese C, Nardini Fm, Orlando S, Perego R, Veneri A
Source: CEUR WORKSHOP PROCEEDINGS. Milano, Italy, 29-30/06/2022
See at:
ceur-ws.org | CNR IRIS | ISTI Repository | CNR IRIS