Overview
This workshop brings together researchers and PhD students to exchange ideas and stimulate collaborations on optimization methods for machine learning and embedded control. The 3-day event features invited talks from leading experts and poster presentations highlighting innovative work by PhD students.
Speakers
Stephen Boyd
Stanford University, USA
Title: TBD
Talk summary.
Volkan Cevher
EPFL, Switzerland
Title: TBD
TBD
Moritz Diehl
University of Freiburg, Germany
Title: TBD
TBD
Mikael Johansson
KTH Royal Institute of Technology, Sweden
Title: TBD
TBD.
Ion Necoara
University Politehnica Bucharest, Romania
Title: Linearized augmented Lagrangian methods for optimization problems with nonlinear equality constraints
We consider (nonsmooth) nonconvex optimization problems with nonlinear equality constraints. We assume that (some part of) the objective function and the functional constraints are locally smooth. For solving this problem, we propose linearized augmented Lagrangian methods, i.e., we linearize the objective function and the functional constraints in a Gauss-Newton fashion at the current iterate within the augmented Lagrangian function and add a quadratic regularization, yielding a (convex) subproblem that is easy to solve, and whose solution is the next primal iterate. The update of the dual multipliers is also based on the linearization of functional constraints. Under some specific dynamic regularization parameter choices, we prove boundedness and global asymptotic convergence of the iterates to a first-order solution of the problem. We also derive convergence guarantees for the iterates to an ϵ-first-order solution.
Yurii Nesterov
Center for Operations Research at Corvinus University of Budapest, Hungary, and
School of Data Science at the Chinese University of Hong Kong, Hong Kong SAR
Title: Asymmetric Long-Step Primal-Dual Interior-Point Methods with Dual Centering
We discuss a new way of development of asymmetric Interior-Point Methods for solving primal-dual problems of Conic Optimization. It is very efficient for problems, where the dual formulation is simpler than the primal one. The problems of this type arise often in Semidefinite Optimization (SDO), for which we propose a new primal-dual method with very attractive computational cost. In this approach, we do not need sophisticated Linear Algebra, restricting ourselves by standard Cholesky factorization. However, our complexity bounds correspond to the best-known polynomial-time results. Moreover, for symmetric cones the bounds automatically depend on the minimal barrier parameter between the primal and the dual feasible sets. We show by SDO-examples that the corresponding gain can be very big. We discuss some classes of SDO-problems, where the complexity bounds are proportional to the square root of the number of linear equality constraints and the computational cost of one iteration is as in Linear Optimization. Our theoretical developments are supported by encouraging numerical testing.
Giuseppe Notarstefano
University of Bologna, Italy
Title: TBD
TBD
Panos Patrinos
KU Leuven, Belgium
Title: TBD
TBD
Poster Presentation
TBD
Organizing Committee
- Alberto Bemporad (IMT School for Advanced Studies Lucca, Italy)
- Mario Zanon (IMT School for Advanced Studies Lucca, Italy)
- Moritz Diehl (University of Freiburg, Germany)
- Panagiotis Patrinos (KU Leuven, Belgium)
- Puya Latafat (IMT School for Advanced Studies Lucca, Italy)
Program
- TBD
Venue
Aula Guinigi, IMT School for Advanced Studies Lucca
Piazza San Francesco, 19, Lucca, 55100 Lucca LU, Italy
Travel
From Pisa Airport (PSA) to Lucca
The easiest way to reach Lucca from Pisa Airport is by train.
- Take the PisaMover: From the airport terminal, take the PisaMover shuttle train to Pisa Centrale Station. The journey takes about 5 minutes, and departures are frequent.
- Train to Lucca: From Pisa Centrale, take a regional train to Lucca. The journey takes approximately 25-30 minutes, and trains run regularly.
You can purchase train tickets at Pisa Centrale or online via the Trenitalia website.
From Florence Airport (FLR) to Lucca
To travel from Florence Airport to Lucca:
- Tram to Florence Station: Take the T2 tram line from Florence Airport to Firenze Santa Maria Novella (SMN) train station. The tram ride is about 20 minutes.
- Train to Lucca: From Firenze SMN, take a regional train to Lucca. The journey typically takes 1 hour 15 minutes to 1 hour 45 minutes, depending on the specific train.
Tickets can be bought at the airport, at Florence SMN, or online.
Getting to IMT Lucca from Lucca Train Station
IMT Lucca is conveniently located within a short, walkable distance from Lucca Train Station. The walk is approximately 15-20 minutes and takes you through the charming streets of Lucca.
Upon exiting the train station, follow the signs towards the city center (centro storico). Continue straight along Via del Pallone, then turn left onto Via Santa Croce. Piazza San Francesco will be on your right.
Contact
For inquiries, please email: eventi@imtlucca.it