Matteo Lapucci

PhD Student of Smart Computing at Dipartimento di Ingegneria dell’Informazione, Università degli Studi di Firenze.

My research interests include constrained, multi-objective and sparse non-linear optimization and optimization methods for machine learning and statistics.

Publications

  1. Pareto front approximation through a multi-objective augmented
    Lagrangian method.
    G. Cocchi, M. Lapucci, P. Mansueto.
    EURO Journal on Computational Optimization (2021).
    DOI: 10.1016/j.ejco.2021.100008
  2. An effective procedure for feature subset selection in logistic regression based on information criteria.
    E. Civitelli, M. Lapucci, F. Schoen, A. Sortino.
    Computational Optimization and Applications (2021).
    DOI: 10.1007/s10589-021-00288-1
  3. A Two-Level Decomposition Framework Exploiting First and Second Order Information for SVM Training Problems.
    G. Galvan, M. Lapucci, C.-J. Lin, M. Sciandrone.
    Journal of Machine Learning Research (2021).
    Link: https://jmlr.org/papers/v22/19-632.html
  4. Comprehensive Disease Control in Systemic Lupus Erythematosus.
    F. Ceccarelli, G. Olivieri, A. Sortino, M. Lapucci, M. Sciandrone et al..
    Seminars in Arthritis and Rheumatism (2021).
    DOI: 10.1016/j.semarthrit.2021.02.005
  5. Convergent Inexact Penalty Decomposition Methods for Cardinality-Constrained Problems.
    M. Lapucci, T. Levato, M. Sciandrone.
    Journal of Optimization Theory and Applications (2020).
    DOI: 10.1007/s10957-020-01793-9
  6. An augmented Lagrangian algorithm for multi-objective optimization.
    G. Cocchi, M. Lapucci.
    Computational Optimization and Applications (2020).
    DOI: 10.1007/s10589-020-00204-z
  7. An efficient optimization approach for best subset selection in linear regression, with application to model selection and fitting in autoregressive time-series.
    L. Di Gangi, M. Lapucci, F. Schoen, A. Sortino
    Computational Optimization and Applications (2019).
    DOI: 10.1007/s10589-019-00134-5
  8. On the convergence of inexact Augmented Lagrangian methods for problems with convex constraints.
    G. Galvan, M. Lapucci
    Operations Research Letters (2019).
    DOI: 10.1016/j.orl.2019.03.006
  9. An Alternating Augmented Lagrangian method for constrained nonconvex optimization.
    G. Galvan, M. Lapucci, T. Levato, M. Sciandrone
    Optimization Methods and Software (2019).
    DOI: 10.1080/10556788.2019.1576177

My ORCID number: 0000-0002-2488-5486

Talks

  1. ODS 2021 (Rome)A Unifying Framework for Sparsity Constrained Optimization
  2. SIMAI 2020+2021 (Parma)A Two-Level Decomposition Framework  Exploiting First and Second Order Information for SVM Training  Problems
  3. SIAM Conference on Optimization (OP21)A Penalty Decomposition Approach for Multi-Objective Cardinality-Constrained Optimization Problems.
  4. EUROPT 2021 (Toulouse, virtual)A Derivative-free Adaptation of the Penalty Decomposition Method for Sparse Optimization.
  5. ODS 2019 (Genova) An Efficient Optimization Approach for Subset Selection, with Application to Linear Regression and Auto-Regressive Time Series.

Contacts

Via di Santa Marta, 3 – 50139 Firenze FI (Italy)

E-mail: matlapucci at gmail dot com