PhD Student in Information Engineering, Department of Information Engineering (DINFO), University of Florence.

My research focuses on optimization techniques for mini-batch optimization and in particular to optimization techniques that are suitable to train neural-networks.

Pucci Davide

Publications

  1. A Globally Convergent Gradient Method with Momentum.
    M. Lapucci, G. Liuzzi, S. Lucidi, D. Pucci, M. Sciandrone.
    Computational Optimization and Applications (2025). DOI: 10.1007/s10589-025-00741-5
  2. On the computation of the efficient frontier in advanced sparse portfolio optimization.
    A. Annunziata, M. Lapucci, P. Mansueto, D. Pucci.
    4OR (2025). DOI: 10.1007/s10288-025-00600-3
  3. Convergence conditions for stochastic line search based optimization of over-parametrized models.
    M. Lapucci, D. Pucci.
    Optimization (2025). DOI: 10.1080/02331934.2025.2551250
  4. Joint-based action progress prediction.
    D. Pucci, F. Becattini, A. Del Bimbo.
    Sensors (2023). DOI: 10.3390/s23010520
  5. Mixed-integer quadratic programming reformulations of multi-task learning models.
    M. Lapucci, D. Pucci.
    Mathematics in Engineering (2023). DOI: 10.3934/mine.2023020

Preprints

  1. Penalty decomposition derivative free method for the minimization of partially separable functions over a convex feasible set.
    F Cecere, M Lapucci, D Pucci, M Sciandrone.
    arXiv pre-print (2025). arXiv:2503.21631
  2. Effectively Leveraging Momentum Terms in Stochastic Line Search Frameworks for Fast Optimization of Finite-Sum Problems.
    M Lapucci, D Pucci.
    arXiv pre-print (2024). arXiv:2411.07102
  3. Effective Front-Descent Algorithms with Convergence Guarantees.
    M. Lapucci, P. Mansueto, D. Pucci.
    arXiv pre-print (2024). arXiv:2405.08450

Contacts

Via di Santa Marta, 3 – 50139 Firenze (FI), Italy
E-mail: davide.pucci(AT)unifi.it