no code implementations • 17 Apr 2024 • Alexander Davydov, Francesco Bullo
Contraction theory is a mathematical framework for studying the convergence, robustness, and modularity properties of dynamical systems and algorithms.
no code implementations • 12 Feb 2024 • Sean Jaffe, Alexander Davydov, Deniz Lapsekili, Ambuj Singh, Francesco Bullo
Global stability and robustness guarantees in learned dynamical systems are essential to ensure well-behavedness of the systems in the face of uncertainty.
no code implementations • 7 Nov 2023 • Veronica Centorrino, Anand Gokhale, Alexander Davydov, Giovanni Russo, Francesco Bullo
Our analysis leverages contraction theory to characterize the behavior of a family of firing-rate competitive networks for sparse reconstruction with and without non-negativity constraints.
no code implementations • 8 Aug 2022 • Saber Jafarpour, Alexander Davydov, Matthew Abate, Francesco Bullo, Samuel Coogan
Third, we use the upper bounds of the Lipschitz constants and the upper bounds of the tight inclusion functions to design two algorithms for the training and robustness verification of implicit neural networks.
1 code implementation • 1 Apr 2022 • Alexander Davydov, Saber Jafarpour, Matthew Abate, Francesco Bullo, Samuel Coogan
We use interval reachability analysis to obtain robustness guarantees for implicit neural networks (INNs).
no code implementations • 10 Dec 2021 • Saber Jafarpour, Matthew Abate, Alexander Davydov, Francesco Bullo, Samuel Coogan
First, given an implicit neural network, we introduce a related embedded network and show that, given an $\ell_\infty$-norm box constraint on the input, the embedded network provides an $\ell_\infty$-norm box overapproximation for the output of the given network.
1 code implementation • NeurIPS 2021 • Saber Jafarpour, Alexander Davydov, Anton V. Proskurnikov, Francesco Bullo
Additionally, we design a training problem with the well-posedness condition and the average iteration as constraints and, to achieve robust models, with the input-output Lipschitz constant as a regularizer.