DECODE at ECC2024, Stockholm, Sweden

© 2024 EPFL

© 2024 EPFL

DECODE organized a successful workshop on 'Neural network control with stability guarantees' with PoliMi, and presented several papers on regret minimization and unconstrained optimization of neural networks for control and system identification.

During the exciting week of the 22nd European Control Conference ECC2024 in Stockholm, Sweden, DECODE organized a successful and well-received workshop on neural network control of nonlinear dynamical systems with stability guarantees. The workshop was organized in collaboration with Marcello Farina and Alessio La Bella from Politecnico di Milano, Ian Manchester from the University of Sydney, and Sophie Tabouriech from CNRS, France. Moreover, several members of DECODE, including Andrea Martin, Daniele Martinelli, Danilo Saccani, Luca Furieri, Leonardo Massai, and Muhammad Zakwan, presented the following papers leading to fruitful discussions and future collaboration.

  1. Martinelli D, Martin A, Ferrari-Trecate G, Furieri L. Closing the Gap to Quadratic Invariance: a Regret Minimization Approach to Optimal Distributed Control. ECC2024, Stockholm, Sweden.
  2. Martin A, Furieri L, Dörfler F, Lygeros J, Ferrari-Trecate G. Regret optimal control for uncertain stochastic systems. European Journal of Control. 2024 Jun 25:101051.
  3. Massai L, Saccani D, Furieri L, Ferrari-Trecate G. Unconstrained learning of networked nonlinear systems via free parametrization of stable interconnected operators. ECC2024, Stockholm, Sweden.
  4. Di Natale L*, Zakwan M*, Svetozarevic B, Heer P, Trecate GF, Jones CN. Stable Linear Subspace Identification: A Machine Learning Approach. ECC2024, Stockholm, Sweden.
Funding

DECODE acknowledges NCCR Automation and SNSF for the funding.