Ruggero G. Pensa
banner
rupensa.bsky.social
Ruggero G. Pensa
@rupensa.bsky.social
Computer engineer, professor, scientist @ University of Turin. #MachineLearning #Privacy #XAI #ResponsibleAI
Reposted by Ruggero G. Pensa
La circolazione dei dati sanitari: un dialogo interdisciplinare sullo spazio europeo dei dati.
📅 30–31 ottobre | 📍 Campus Luigi Einaudi – Torino
con una sessione moderata da Ruggero Pensa e un intervento di Alessio Famiani (#UniTo).
👉 www.dg.unito.it/do/a...
October 28, 2025 at 12:00 PM
Now online: "Combining SHAP-Driven Co-clustering and Shallow Decision Trees to Explain XGBoost", presented during the last Discovery Science conference held Pisa in October. #xai #clustering #MachineLearning @diunito.bsky.social @insalyon.bsky.social

link.springer.com/chapter/10.1...
Combining SHAP-Driven Co-clustering and Shallow Decision Trees to Explain XGBoost
Transparency is a non-functional requirement of machine learning that promotes interpretable or easily explainable outcomes. Unfortunately, interpretable classification models, such as linear, rule-ba...
link.springer.com
January 28, 2025 at 8:18 AM
I’m happy to announce that our paper on Differentially Private Associative Co-Clustering has been accepted for #SIAMSDM25! Se you in Alexandria, VA in May 2025. @diunito.bsky.social
December 21, 2024 at 6:04 PM
We are happy to inform you that your submission was accepted to appear at #SDM 2025. 🥳 #SIAM #DataMining
December 20, 2024 at 9:27 PM
... And I can't forget our nice work about our educational activities in schools to help pupils recognize #FakeNews: ieeexplore.ieee.org/document/105....
It is the latest achievement within our successfull project on #SocialMedia #education in #schools (www.social4school.net) @diunito.bsky.social.
Social4school
www.social4school.net
December 11, 2024 at 2:17 PM
[3/3] The last paper is one of the results of my visiting period at #Inria Lyon, during my sabbatical leave. It's about explaining #XGBoost by combining #co-clustering, #SHAP and shallow decision trees. I presented it in Pisa, at #DiscoveryScience (ds2024.isti.cnr.it).
#XAI #explainability
Discovery Science 2024
ds2024.isti.cnr.it
December 11, 2024 at 2:09 PM
[2/3] The second one presents a method for learning #fair distances among categorical data. It's effective in different #MachineLearning tasks for tabular data, including #classification and #clustering. It has been presented in the BIAS workshop @ecmlpkdd.org 2024:
drive.google.com/file/d/1d7oD...
Interpretable Fair Distance Learning for Categorical Data.pdf
drive.google.com
December 11, 2024 at 1:55 PM
[1/3] As 2024 is heading to an end, let me mention a couple of research papers authored by me and my group in the last year.
Let's start with the first one, a #survey on #co-clustering that also includes an experimental comparison of the main methods: dl.acm.org/doi/10.1145/...
#ai #MachineLearning
Co-clustering: A Survey of the Main Methods, Recent Trends, and Open Problems | ACM Computing Surveys
Since its early formulations, co-clustering has gained popularity and interest both within and outside the machine learning community as a powerful learning paradigm for clustering high-dimensional da...
dl.acm.org
December 11, 2024 at 1:42 PM