Rome Workshop on Veridical Data Science

A one day workshop on Veridical Data Science (VDS) in Rome, is co-organized by Integreat, the Norwegian centre for knowledge-driven machine learning in Oslo, the Department of Statistics, University of California, Berkeley, the L’EMbeDS Department of Excellence at the Sant’Anna School of Advanced Studies at Pisa, and the Department of Methods and Models for Economics, Territory and Finance of the University of Roma La Sapienza and the Institute for Systems Analysis and Computer Science of the Italian National Research Council.

Time and place: June 20, 2025 9:00 AM–6:00 PM, Aula Acquario, MEMOTEF, University of Roma la Sapienza, Rome,

About

This veridical data science one-day workshop at University of Roma la Sapienza, Rome, aims at showcasing and promoting veridical (truthful) data science (VDS) towards reproducible and reliable data analysis and decisions. It intends to build a community of VDS researchers in Europe for trustworthy data science and machine learning. The workshop discussions will provide opportunities for statisticians and data scientists to identify important VDS research topics and critical applications in academia and industry. Graduate students and early career researchers will benefit from this workshop to find future research directions.

Previous workshops

The first VDS workshop was held at UC Berkeley on May 31, 2024: Inaugural Berkeley-Stanford Workshop on Veridical Data Science (talk videos available)

Call for oral contributions

Submission is closed.

REGISTRATION

Registration is required and is free. Registration form.
Deadline for registration: 18 June 2025.

PROGRAMME

9:00 – 9:15	Brunero Liseo (Roma La Sapienza), Arnoldo Frigessi (Oslo)	Welcome and introduction
9:15 – 10:15	Bin Yu (UC Berkeley)	Veridical data science and PCS uncertainty quantification
10:15 – 10:50	Samuel Pawel, (Center for Reproducible Science, University of Zurich)	Meta-scientific perspectives on simulation studies in methodological research: Status quo, issues, and potential solutions
10:50 – 11:10	Break
11:10 – 11:45	Giovanni Felici (IIASI Roma, CNR)	From Optimization to Learning, and back
11:45 – 12:00	Riccardo Russo (Department of Economics and Statistics, Banca d’Italia)	Satellite-Based Data For CO2 Monitoring. Potential And Limits
12:00 – 12:15	Giulia Iadisernia (Department of Information Technology, Banca d’Italia)	Prompting for policy: Scaling reproducible prompt evaluation for economics research
12:15 – 12:30	Jules Perez (University of California, Berkeley)	Assessment of genetic diversity metrics for conservation of diverse plant species
12:30 – 13:30	Lunch
13:30 – 14:05	Dino Pedreschi (University of Pisa)	Human-AI coevolution
14:05 – 14:40	Stéphane Guerrier (University of Geneva)	A Step Toward More Replicable Studies: Charting and Combining Forking Paths
14:40 – 14:55	Julien Randon-Furling (Centre Borelli, ENS Paris-Saclay, Université Paris-Saclay Alliance)	Reliable Agent-Based Models for Decision and Policy Making
14:55 – 15:10	Gian Mario Sangiovanni (Department of Statistical Sciences, University of Rome La Sapienza)	From Object Detection to Modeling Spatial Intensity of Marine Biodiversity: Joining Machine Learning and Bayesian Statistics
15:10 – 15:30	Break
15:30 – 16:05	Clelia di Serio (Università Vita-Salute San Raffaele, Milano)	Data sharing attitude towards reproducibility in biomedical research: nature or nurture? A statistical perspective
16:05 – 16:40	Roberto Molinari (Auburn University)	A Rashomon Algorithm for Sparse Model Sets
16:40 - 17:30	Bin Yu, Francesca Chiaromonte, Clelia Di Serio, Brunero Liseo, Samuel Pawel, Arnoldo Frigessi	Panel discussion and conclusions

Confirmed speakers:

Bin Yu (UC Berkeley): Veridical data science and PCS uncertainty quantification
Clelia di Serio (Università Vita-Salute San Raffaele, Milano): Data sharing attitude towards reproducibility in biomedical research: nature or nurture? A statistical perspective.
Samuel Pawel (Center for Reproducible Science, University of Zurich): Meta-scientific perspectives on simulation studies in methodological research: Status quo, issues, and potential solutions.
Dino Pedreschi (University of Pisa): Human-AI coevolution
Stéphane Guerrier (University of Geneva): A Step Toward More Replicable Studies: Charting and Combining Forking Paths
Giovanni Felici (CNR-IASI Roma): From Optimization to Learning, and back
Roberto Molinari (Auburn University): A Rashomon Algorithm for Sparse Model Sets

Confirmed short contributed presentations:

Riccardo Russo (Department of Economics and Statistics, Banca d’Italia): Satellite-Based Data For CO2 Monitoring. Potential And Limits
Giulia Iadisernia (Department of Information Technology, Banca d’Italia): Prompting for policy: Scaling reproducible prompt evaluation for economics research
Jules Perez (University of California, Berkeley): Assessment of genetic diversity metrics for conservation of diverse plant species
Julien Randon-Furling (Centre Borelli, ENS Paris-Saclay, Université Paris-Saclay Alliance): Reliable Agent-Based Models for Decision and Policy Making
Gian Mario Sangiovanni (Department of Statistical Sciences, University of Rome La Sapienza): From Object Detection to Modeling Spatial Intensity of Marine Biodiversity: Joining Machine Learning and Bayesian Statistics

Organisers:

Brunero Liseo, MEMOTEF, Roma

Arnoldo Frigessi, Integreat, University of Oslo

Bin Yu, Department of Staistics, University of California, Berkeley

Francesca Chiaromonte, Sant’Anna School of Advanced Studies, Pisa; Huck Institutes of the Life Sciences, The Pennsylvania State University

For information, please use this contact admin@integreat.no.

Organiser

Integreat, the Norwegian Centre for Knowledge-driven Machine Learning, Department of Methods and Models for Economics, Territory and Finance, University of Roma La Sapienza, Department of Statistics, University of California, Berkeley, Sant’Anna School of Advanced Studies at Pisa and IASI – Institute for System Analysis and Computer Science

Published Mar. 5, 2025 10:33 AM - Last modified June 17, 2025 10:38 AM