AIDA Lab

Algorithms for Intelligent Data Analytics

We are a research group in the Department of Information Engineering of University of Padova.

Our research is focused on the development of algorithms for extracting useful information from large datasets. Our work spans areas such as theory of computation, data mining, machine learning with applications to the analysis of big data from social networks, biomedical data, urban traffic, industrial processes.

We focus on algorithms for: modern computing architectures; pattern mining and learning; temporal, dynamic, and evolving data; unsupervised learning (clustering, diversity).

The common themes of our work include scalability, fairness and privacy, networked data, and rigorous guarantees.

People

Faculty

• Matteo Ceccarello (Assistant Professor)
• Leonardo Pellegrina (Assistant Professor)
• Andrea Alberto Pietracaprina (Full Professor)
• Geppino Pucci (Full Professor)
• Francesco Silvestri (Associate Professor)
• Fabio Vandin (Full Professor)

Postdocs and PhD students

• Cristian Boldrin (Ph.D student)
• Fabrizio Boninsegna (Phd student)
• Antonio Collesei (Phd student)
• Dario Simionato (PhD student)
• Lorenzo Padoan (Research collaborator)
• Giorgio Venturin (Ph.D student)

• Davide Buffelli (now @ MediaTek Research)
• Andrea Tonon (now @ Huawei)
• Diego Santoro
• Ilie Sarpe (now @ KTH)

Recent Publications

2023

• Leonardo Pellegrina, Fabio Vandin SILVAN: Estimating Betweenness Centralities with Progressive Sampling and Non-uniform Rademacher Bounds ACM Trans. Knowl. Discov. Data
• Andrea Tonon, Fabio Vandin caSPiTa: mining statistically significant paths in time series data from an unknown network. Knowl. Inf. Syst.
• Dario Simionato, Fabio Vandin Bounding the Family-Wise Error Rate in Local Causal Discovery Using Rademacher Averages (Extended Abstract). IJCAI
• Enrico Dandolo, Andrea Pietracaprina, Geppino Pucci Distributed k-Means with Outliers in General Metrics. Euro-Par
• Leonardo Pellegrina Efficient Centrality Maximization with Rademacher Averages. KDD
• Matteo Ceccarello, Andrea Pietracaprina, Geppino Pucci, Federico Soldà Scalable and space-efficient Robust Matroid Center algorithms. J. Big Data
• Matteo Ceccarello, Anton Dignös, Johann Gamper, Christina Khnaisser Indexing Temporal Relations for Range-Duration Queries. SSDBM
• Paolo Pellizzoni, Andrea Pietracaprina, Geppino Pucci Fully Dynamic Clustering and Diversity Maximization in Doubling Metrics. WADS
• Paolo Pellizzoni, Fabio Vandin VC-dimension and Rademacher Averages of Subgraphs, with Applications to Graph Mining. ICDE
• Shiyuan Deng, Francesco Silvestri , Yufei Tao Enumerating Subgraphs of Constant Sizes in External Memory. ICDT

2022

• A. Guiotto, G. Bortolami, A. Ciniglio, F. Spolaor, G. Guarneri, A. Avogaro, F. Cibin, F. Silvestri, Z. Sawacha. Machine learning approach to diabetic foot risk classification with biomechanics data Gait & Posture
• Adam Charane, Matteo Ceccarello, Anton Dignös, Johann Gamper Efficient Computation of All-Window Length Correlations. DB&IS
• Andrea Tonon, Fabio Vandin gRosSo: mining statistically robust patterns from a sequence of datasets. Knowl. Inf. Syst.
• Dario Simionato, Fabio Vandin Bounding the Family-Wise Error Rate in Local Causal Discovery Using Rademacher Averages. ECML/PKDD
• Davide Buffelli, Fabio Vandin The Impact of Global Structural Information in Graph Neural Networks Applications. Data
• Davide Buffelli, Fabio Vandin Graph Representation Learning for Multi-Task Settings: a Meta-Learning Approach. IJCNN
• Davide Buffelli, Pietro Lió, Fabio Vandin SizeShiftReg: a Regularization Method for Improving Size-Generalization in Graph Neural Networks NeurIPS
• Davide Buffelli, Pietro Lió, Fabio Vandin SizeShiftReg: a Regularization Method for Improving Size-Generalization in Graph Neural Networks. NeurIPS
• Diego Santoro, Ilie Sarpe ONBRA: Rigorous Estimation of the Temporal Betweenness Centrality in Temporal Networks. WWW
• Diego Santoro, Leonardo Pellegrina, Matteo Comin, Fabio Vandin SPRISS: approximating frequent k-mers by sampling reads, and applications. Bioinform.
• Fabio Vandin Technical perspective: Evaluating sampled metrics is challenging. Commun. ACM
• Johann Gamper, Matteo Ceccarello, Anton Dignös What's New in Temporal Databases? ADBIS
• Leonardo Pellegrina, Cyrus Cousins, Fabio Vandin, Matteo Riondato MCRapper: Monte-Carlo Rademacher Averages for Poset Families and Approximate Pattern Mining. ACM Trans. Knowl. Discov. Data
• Leonardo Pellegrina, Fabio Vandin Discovering significant evolutionary trajectories in cancer phylogenies Bioinformatics
• Martin Aumüller , Matteo Ceccarello Implementing Distributed Similarity Joins using Locality Sensitive Hashing. EDBT
• Martin Aumüller , Sariel Har-Peled, Sepideh Mahabadi, Rasmus Pagh, Francesco Silvestri Sampling near neighbors in search for fairness. Commun. ACM
• Martin Aumüller , Sariel Har-Peled, Sepideh Mahabadi, Rasmus Pagh, Francesco Silvestri Sampling a Near Neighbor in High Dimensions - Who is the Fairest of Them All? ACM Trans. Database Syst.
• Matteo Ceccarello, Johann Gamper Fast and Scalable Mining of Time Series Motifs with Probabilistic Guarantees. Proc. VLDB Endow.
• Paolo Pellizzoni, Andrea Pietracaprina, Geppino Pucci k-Center Clustering with Outliers in Sliding Windows. Algorithms
• Paolo Pellizzoni, Andrea Pietracaprina, Geppino Pucci Adaptive k-center and diameter estimation in sliding windows. Int. J. Data Sci. Anal.
• Paolo Sylos Labini, Massimo Bernaschi, Werner Nutt, Francesco Silvestri , Flavio Vella Blocking Sparse Matrices to Leverage Dense-Specific Multiplication. IA3@SC

2021

• Andrea Tonon, Fabio Vandin CASPITA: Mining Statistically Significant Paths in Time Series Data from an Unknown Network. ICDM
• Davide Buffelli, Fabio Vandin Attention-Based Deep Learning Framework for Human Activity Recognition With User Adaptation IEEE Sensors
• Diego Santoro, Leonardo Pellegrina, Fabio Vandin SPRISS: Approximating Frequent k-mers by Sampling Reads, and Applications RECOMB
• Elia Costa, Francesco Silvestri On the Bike Spreading Problem. ATMOS
• Federico Altieri, Andrea Pietracaprina, Geppino Pucci, Fabio Vandin Scalable Distributed Approximation of Internal Measures for Clustering Evaluation. SDM
• Ilie Sarpe, Fabio Vandin odeN: Simultaneous Approximation of Multiple Motif Counts in Large Temporal Networks. CIKM
• Ilie Sarpe, Fabio Vandin PRESTO: Simple and Scalable Sampling Techniques for the Rigorous Approximation of Temporal Motif Counts. SDM
• Martin Aumüller , Matteo Ceccarello The role of local dimensionality measures in benchmarking nearest neighbor search. Inf. Syst.
• Martin Aumüller , Sariel Har-Peled, Sepideh Mahabadi, Rasmus Pagh, Francesco Silvestri Fair near neighbor search via sampling. SIGMOD Rec.
• Matteo Comin, Barbara Di Camillo, Cinzia Pizzi, Fabio Vandin Comparison of microbiome samples: methods and computational challenges. Briefings Bioinform.
• Rezaul Chowdhury, Francesco Silvestri , Flavio Vella Algorithm Design for Tensor Units. Euro-Par

2020

• Andrea Tonon, Fabio Vandin GRosSo: Mining Statistically Robust Patterns from a Sequence of Datasets. ICDM
• Diego Santoro, Andrea Tonon, Fabio Vandin Mining Sequential Patterns with VC-Dimension and Rademacher Complexity. Algorithms
• Leonardo Pellegrina, Cinzia Pizzi, Fabio Vandin Fast Approximation of Frequent k-Mers and Applications to Metagenomics. J. Comput. Biol.
• Leonardo Pellegrina, Cyrus Cousins, Fabio Vandin, Matteo Riondato MCRapper: Monte-Carlo Rademacher Averages for Poset Families and Approximate Pattern Mining. KDD
• Leonardo Pellegrina, Fabio Vandin Efficient mining of the most significant patterns with permutation testing. Data Min. Knowl. Discov.
• Martin Aumüller , Matteo Ceccarello Running Experiments with Confidence and Sanity. SISAP
• Martin Aumüller , Rasmus Pagh, Francesco Silvestri Fair Near Neighbor Search: Independent Range Sampling in High Dimensions. PODS
• Matteo Ceccarello, Andrea Pietracaprina, Geppino Pucci A General Coreset-Based Approach to Diversity Maximization under Matroid Constraints. ACM Trans. Knowl. Discov. Data
• Matteo Ceccarello, Andrea Pietracaprina, Geppino Pucci, Eli Upfal Distributed Graph Diameter Approximation. Algorithms
• Matteo Riondato, Fabio Vandin MiSoSouP: Mining Interesting Subgroups with Sampling and Pseudodimension. ACM Trans. Knowl. Discov. Data
• Paolo Pellizzoni, Andrea Pietracaprina, Geppino Pucci Dimensionality-adaptive k-center in sliding windows. DSAA
• Rezaul Chowdhury, Francesco Silvestri , Flavio Vella A Computational Model for Tensor Core Units. SPAA
• Thomas D. Ahle, Francesco Silvestri Similarity Search with Tensor Core Units. SISAP
• Yoo-Ah Kim, D. Wojtowicz, R. Sarto Basso, I. Sason, W. Robinson, D. S. Hochbaum, M. Leiserson, R. Sharan, F. Vandin, T. Przytycka. Network-based approaches elucidate differences within APOBEC and clock-like signatures in breast cancer Genome Medicine

Recent Projects

• [2023-2025] Ministero dell’Università e della Ricerca (MUR), PRIN EXPAND: scalable algorithms for EX- Ploratory Analyses of heterogeneous and dynamic Networked Data Role: PI. Total amount: euro 388,337.
• [2023-2024] University of Padova, Big-Mobility: Big-data Analytics for Mobility. Role: PI. Total amount: euro 68,000.
• [2022-2025] Ministero dell’Università e della Ricerca (MUR), National Research Centre for High Performance Computing, Big Data and Quantum Computing. Role: participant Total amount: euro 763,986.
• [2021-2024] European Union, Brainteaser: BRinging Artificial INTelligencE home for a better cAre of amyotrophic lateral sclerosis and multiple SclERosis. Role: participants. Total UniPD amount: euro 732,250.
• [2019-2023] Ministero dell’Istruzione, dell’Università e della Ricerca (MIUR), PRIN AHeAD: efficient Algorithms for HArnessing networked Data. Role: co-PI. Total amount: euro 784,860.
• [2018-2020] University of Padova, STARS: Algorithms for Inferential Data Mining. Role: PI. Total amount: euro 140,000.
• [2017-2019] University of Padova, Algorithms for Networks Analysis and Bioinformatics Applications. Role: PI. Total amount: euro 54,000.
• [2017-2020] University of Padova, From Single-Cell to Multi-Cells Information Systems Analysis. Role: co-PI. Total amount: euro 220,000.
• [2012-2018] National Science Foundation, BIGDATA: Mid-Scale: Analytical Approaches to Massive Data Computation with Applications to Genomics. Role: co-PI Total amount: $1,566,685.

Awards

• Test of Time award, RECOMB 2023 (Conference on Research in Computational Molecular Biology 2023)
• Best Paper Award, EURO-PAR 2023 (29th International European Conference of Parallel and Distributed Computing)
• Best Paper Award, SSDBM 2023 (35th International Conference on Scientific and Statistical Database Management)
• Best Paper Award, ICDT 2023 (26th International Conference on Database Theory)
• Best Paper Award, ECML-PKDD 2022 (European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2022)
• Runner-up for Test of Time award, RECOMB 2022 (Conference on Research in Computational Molecular Biology 2022)
• Invited articles to the Communication of the ACM and to SIGMOD Records for the ACM SIGMOD Research Highlights Award
• ACM SIGMOD Research Highlights Award 2021, awarded for a paper at ACM PODS 2020 (Symposium on Principles of Database Systems)
• Invited article to the special issue of KAIS for the best papers of IEEE ICDM 2020 (International Conference on Data Mining)
• Best Paper Award of SISAP 2019 (International Conference on Similarity Search and Applications)
• Runner up for the Facebook AI System Hardware/Software Co-Design research awards 2019
• Invited article to the special issue of ACM TKDD for the best papers of ACM KDD 2018 (International Conference on Knowledge Discovery and Data Mining)
• Invited article to the special issue of Algorithms for Molecular Biology for the best papers of WABI 2018 (Workshop on Algorithms for Bioinformatics)
• Best Paper Award, RECOMB 2013 (Conference on Research in Computational Molecular Biology 2013)
• Invited article to the special issue of Algorithms for Molecular Biology for the best papers of WABI 2011 (Workshop on Algorithms for Bioinformatics)
• Invited article to the special issue of DAMI for the best papers of ECML-PKDD 2010 (European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2010)
• Best Paper Award, Algorithms Track, IEEE IPDPS 2010 (24th International Parallel and Distributed Processing Symposium)