She received her Master degree in Electronic Engineering summa cum laude in 1990 and her Ph.D. degree in Operations Research in 1995 from the Sapienza University of Roma. She has been a professor in the School of since 2000 and a faculty member at DIS since 1996.
Palagi’s research focus is mainly on continuous optimization. Among the main research interest:
- Algorithm for the solution of special class of Semidefinite Programming Problems (SDP);
- Decomposition methods for constrained problems arising in training Support Vector Machines;
- Solution Techniques for Standard Quadratic Problems (StQP);
- Algorithms for the solution of quadratrically constrained problems
- Local algorithm for general nonlinear programming Problems (NLP);
- Continuous differentiable Merit functions
- Global convergent for general nonlinear programming Problems (NLP).
More recent research topics are Mixed intenger Nonlinear Problems (MINLP) and definition ofBranch and Bound schemes using SDP bound for MaxCut.
She teaches basic courses on Operations Research and Optimization for BS and advaced course for MS programs. She also organizes and teaches courses in PhD programs.
- L. Grippo, L. Palagi, V. Piccialli, M. Piacentini, G. Rinaldi (2009). SpeeDP: A new algorithm to compute the SDP relaxations of Max-Cut for very large graphs. Accepted for publication in Mathematical Programming, series B
- L. Grippo, L. Palagi, V. Piccialli. An unconstrained minimization method for solving low rank SDP relaxations of the max cut problem. Mathematical Programming, Ser. A.. Volume 126, Number 1, 119-146
- S. Lucidi, L. Palagi, A. Risi, M. Sciandrone. On the convergence of hybrid decomposition methods for SVM training. IEEE Transactions on Neural Networks, Volume 20, Issue 6, June 2009, Page(s):1055 - 1060.
- G. Di Pillo, S. Lucidi, L. Palagi (2005). Convergence to 2-nd order stationary points of primal-dual algorithm models for nonlinear programming.Mathematics of Operations Research, vol.30. No. 4, 2005, pp. 897-015.
- I. Bomze, L. Palagi (2005). Quartic formulation of standard quadratic optimization. Journal of Global Optimization Volume: 32, Issue: 2, pp. 181 - 205.
- C.-J. Lin, S. Lucidi, L. Palagi, A. Risi, M. Sciandrone. A decomposition algorithm model for singly linearly constrained problems subject to lower and upper bounds. J. Optim. Theory Appl. vol.~141, No. 1, pp. 107-126, 2009.
Continuous Optimization, Mixed integer nonlinear programming problems, Semidefinite Programming Problems, Support Vector Machines