Topix
A new copyrighted technology for co-clustering of interaction data involving text and images
About me
Starting from September 2022, I am Junior Professor in AI for Archaeology and History at Université Côte d'Azur. In collaboration with the researchers of the CEPAM laboratory, my job consists of developing new AI models and algorithms to answer research questions related with the historical/archaeological data. Indeed, this data is very scarse, fragmented and sparse. This is why standard machine and deep learning tools are either not adapted or at least not working out of the box. I am also part of the INRIA research team MAASAI in Sophia-Antipolis as well as member of the LJAD in Nice. I previously worked as researcher at the Center of Modeling, Simulation and Interactions of Université Côte d'Azur (Nice, France). I obtained my PhD in November 2017 at the Université Paris 1 Panthéon-Sorbonne, where I was part of the SAMM laboratory. In 2023 I co-organized the first edition of IAMAHA and I am part of the organizing committee of Statlearn since six years.
My research interests include:
- Static and dynamic network analysis : spectral clustering, stochastic block models (SBM), hierarchical random graph models (HRG), latent position models (LPM), graph neural networks.
- Time series analysis : change points detection, hidden Markov models (HMM).
- Exact and approximate inference: stochastic and variational EM algorithms, MCMC.
- Image analysis: generative models and convolutional neural networks.
New (!!)
- Out paper A Deep Dynamic Latent Block Model for Co-clustering of Zero-Inflated Data Matrices (Marchello et al.2024) appeared on JCSG
- Our paper Clustering by Deep Latent Position Model with Graph Convolutional Networks (Liang et al. 2024) appeared on ADAC
- New job offer ( internship + potential PhD)
- Our paper A Deep Dynamic Latent Block Model for the Co-clustering of Zero-Inflated Data Matrices (Marchello et al. 2023) has been accepted at ECML/PKDD 2023
- Our paper Continuous Latent Position Models for Instantaneous Interactions (R. Rastelli and M. Corneli, 2023) has been accepted for publication on Network Science
PhD Students
- Davide Adamo (2023 - ongoing, with E. Vila and M. Vuillien)
- Seydina Ousmane Niang (2023 - ongoing, with C. Bouveyron and P. Latouche)
- Giulia Marchello (2020-2023, with C. Bouveyron)
- Cédric Vincent-Cuaz (2020-2023, with R. Flamary)
- Dingge Liang (2019-2022, with C. Bouveyron and P. Latouche)
Publications
Preprints
- A. Zugarini, E. Meloni, A. Betti, A. Panizza, M. Corneli, M. Gori. An optimal control approach to learning in SIDARTHE epidemic model (2020): [ArXiv]
Book Chapters
- L. Vanni, M. Corneli, D. Longree, D. Mayaffre and F. Precioso. Key passages : from statistics to deep learning , in D.F. Iezzi, D. Mayaffre, M. Misuraca, Text Analytics - Advances and Challenges, Springer, Ch.4, (2020): [Web]
Journal papers
- D. Liang, M. Corneli, C. Bouveyron and P. Latouche. Clustering by Deep Latent Position Model with Graph Convolutional Network , Advances in Data Analysis and Classification (2024): [HAL, Journal link]
- G. Marchello, M. Corneli, C. Bouveyron. A Deep Dynamic Latent Block Model for Co-clustering of Zero-Inflated Data Matrices , Journal of Computational and Graphical Statistics , (2024), Taylor and Francis [Journal link]
- D. Liang, M. Corneli, C. Bouveyron, P. Latouche. The graph embedded topic model , Neurocomputing, 562 (2023) [HAL, Journal link]
- R. Romero, J. Lijttijt, R. Rastelli, M. Corneli, T. De Bie. Gaussian Embedding of Temporal Networks , IEEE Access (2023) [Journal link]
- Rastelli, R., & Corneli, M. (2023). Continuous latent position models for instantaneous interactions , Network Science , 1-29. [ArXiv, Journal link]
- L. Vanni, M. Corneli, D. Mayaffre, F. Precioso. From text saliency to linguistic objects: learning linguistic interpretable markers with a multi-channels convolutional architecture , Corpus , Vol. 24 (online, 2023) [Journal link]
- Marchello, G., Fresse, A., Corneli, M., C. Bouveyron. Co-clustering of evolving count matrices with the dynamic latent block model: application to pharmacovigilance , Statistics and Computing 32, Article number: 41, Springer (2022)[ Journal link ].
- D. Liang, M. Corneli, C. Bouveyron, P. Latouche. DeepLTRS: A deep latent recommender system based on user ratings and reviews , Pattern Recognition Letters, Vol. 152, pp. 267-274, Elsevier (2021): [Journal link]
- M. Corneli, C. Bouveyron, P. Latouche. Co-Clustering of ordinal data via latent continuous random variables and a classification EM algorithm , Journal of Computational and Graphical Statistics, Taylor & Francis (2020): [HAL, Journal link]
- L. Bergé, C. Bouveyron, M. Corneli, P. Latouche. The Latent topic block model for the co-clustering of textual interaction data, Computational Statistics and Data Analysis, vol. 137, pp.247-270, Elsevier (2019): [HAL, Journal link ].
- M. Corneli, C. Bouveyron, P. Latouche, F. Rossi. The dynamic stochastic topic block model for dynamic networks with textual edges , Statistics and Computing , pp.1-19, Springer (2018): [HAL , Journal link].
- M. Corneli, P. Latouche, F. Rossi. Multiple change points detection and clustering in dynamic networks. Statistics and Computing, vol. 28, Issue 5, pp 989–1007, Springer (2018): [HAL , Journal link ].
- M. Corneli, P. Latouche, F. Rossi. Exact ICL maximization in a non-stationary temporal extension of the stochastic block model for dynamic graphs. Neurocomputing , vol. 192, pp.81-91, Elsevier (2016): [HAL , Journal link ].
- M. Corneli, P. Latouche, F. Rossi. Block modelling in dynamic networks with non-homogeneous Poisson processes and exact ICL. Social Network Analysis and Mining , pp. 6:55, Springer (2016): [HAL , Journal link ].
Conference papers
- G. Marchello, M. Corneli & C. Bouveyron. A Deep Dynamic Latent Block Model for Co-clustering of Zero-Inflated Data Matrices , ECML-PKDD 2023: [HAL]
- C. Vincent-Cuaz, T. Vayer, R. Flamary, M.Corneli, N. Courty. Template based Graph Neural Network with Optimal Transport Distances , NeurIPS 2022 : [ArXiv]
- M. Corneli, E. Erosheva. A Bayesian approach for clustering and exact finite-sample model selection in longitudinal data mixtures, ISBA world meeting 2022 : [HAL]
- C. Vincent-Cuaz, T. Vayer, R. Flamary, M.Corneli, N. Courty. Semi-relaxed Gromov-Wasserstein divergence with applications on graphs, ICLR 2022 : [ArXiv]
- C. Vincent-Cuaz, T. Vayer, R. Flamary, M.Corneli, N. Courty. Online Graph Dictionary Learning , ICML 2021 : [HAL]
- G. Marchello, A. Fresse, M. Corneli, C. Bouveyron. Co-clustering of evolving count matrices in pharmacovigilance with the dynamic latent block model, ICLR 2021, Workshop AI for Public Health , May 2021, Virtual Conference (formerly Vienna), Austria. [HAL]
- D. Liang, M. Corneli, P. Latouche, C. Bouveyron. Missing rating imputation based on product reviews via deep latent variable models, ICML Workshop Artemiss (2020) [pdf]
- G. Marchello, M. Corneli, C. Bouveyron. The dynamic latent block model for sparse and evolving count matrices, ICML Workshop Artemiss (2020) [pdf]
- L. Vanni, M. Corneli, D. Longree, D. Mayaffre and F. Precioso. Hyperdeep : deep learning descriptif pour l’analyse de données textuelles,, 15èmes Journées Internationales d’Analyse statistique des Données Textuelles, online proceedings, JADT (2020)
Talks/Posters
Workshop Laboratory LJK, Grenoble, France, Apr 2018.
13th International Conference on Operation Research, Havana, Cuba, March 2018.
Journées YSP (Young Statisticians and Probabilists), Paris, France, Jan 2018.
The 6th International Conference on Complex Networks and Their Applications, Lyon, France, Nov/Dec 2017.
49èmes Journées de Statistique (JdS), Avignon, France, Mai/Juin 2017.
Statlearn, Vannes, France, Apr 2016 (poster).
International Conference on Advances in Social Networks Analysis and Mining (ASONAM), Paris, France Aug 2015 (poster).
47èmes Journées de Statistique (JdS), Lille, France, Juin 2015.
49èmes Journées de Statistique (JdS), Avignon, France, Mai/Juin 2017.
European Symposium on Artificial Neural Networks, Computational Intelligence and ML (ESANN), Bruges, Belgium, Apr 2015.
Teaching
M2 MA IM UniCA (since 2023):
- Statistical learning with and on graphs
- Ch.1 [ notes , Colab1, Colab2]
- Ch.2 [ notes ]
- Ch.3 [ notes , R Markdown]
- Ch.4 [ notes , R Markdown]
MSc Data Science UniCA (2018-2022):
- Basic algebra for data analysis [Lecture notes].
- Statistical inference.
Licence MIASHS Université Paris 1 (2014-2017):
- Linear algebra.
- Statistics and probability.
Ongoing projects
CpDyna
An R/C++ package for change point detection in dynamic graphs (joint work with Margot Selosse)
Software
ordinalLBM: an R package to perform co-clustering of ordinal data (CRAN)
CpDyna: an R/C++ package for change point detection in dynamic graphs (github)
Get In Touch
How to reach me: