Litcius/Paper detail

An Incremental Input-to-State Stability Condition for a Class of Recurrent Neural Networks

William D’Amico, Alessio La Bella, Marcello Farina

2023IEEE Transactions on Automatic Control26 citationsDOIOpen Access PDF

Abstract

This paper proposes a novel sufficient condition for the incremental input-to-state stability of a class of recurrent neural networks (RNNs). The established condition is compared with others available in the literature, showing to be less conservative. Moreover, it can be applied for the design of incremental input-to-state stable RNN-based control systems, resulting in a linear matrix inequality constraint for some specific RNN architectures. The formulation of nonlinear observers for the considered system class, as well as the design of control schemes with explicit integral action, are also investigated. The theoretical results are validated through simulation on a referenced nonlinear system.

Topics & Concepts

Recurrent neural networkControl theory (sociology)Nonlinear systemConstraint (computer-aided design)Class (philosophy)Stability (learning theory)Artificial neural networkComputer scienceState (computer science)MathematicsMathematical optimizationControl (management)Artificial intelligenceAlgorithmMachine learningQuantum mechanicsPhysicsGeometryControl and Stability of Dynamical SystemsModel Reduction and Neural NetworksNeural Networks and Applications
An Incremental Input-to-State Stability Condition for a Class of Recurrent Neural Networks | Litcius