RobustNeuralNetworks.jl: a Package for Machine Learning and Data-Driven Control with Certified Robustness

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Barbara, Nicholas H., Revay, Max, Wang, Ruigang, Cheng, Jing, Manchester, Ian R.
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915523166470144
author Barbara, Nicholas H.
Revay, Max
Wang, Ruigang
Cheng, Jing
Manchester, Ian R.
author_facet Barbara, Nicholas H.
Revay, Max
Wang, Ruigang
Cheng, Jing
Manchester, Ian R.
contents Neural networks are typically sensitive to small input perturbations, leading to unexpected or brittle behaviour. We present RobustNeuralNetworks.jl: a Julia package for neural network models that are constructed to naturally satisfy a set of user-defined robustness metrics. The package is based on the recently proposed Recurrent Equilibrium Network (REN) and Lipschitz-Bounded Deep Network (LBDN) model classes, and is designed to interface directly with Julia's most widely-used machine learning package, Flux.jl. We discuss the theory behind our model parameterization, give an overview of the package, and provide a tutorial demonstrating its use in image classification, reinforcement learning, and nonlinear state-observer design.
format Preprint
id arxiv_https___arxiv_org_abs_2306_12612
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle RobustNeuralNetworks.jl: a Package for Machine Learning and Data-Driven Control with Certified Robustness
Barbara, Nicholas H.
Revay, Max
Wang, Ruigang
Cheng, Jing
Manchester, Ian R.
Machine Learning
Systems and Control
Neural networks are typically sensitive to small input perturbations, leading to unexpected or brittle behaviour. We present RobustNeuralNetworks.jl: a Julia package for neural network models that are constructed to naturally satisfy a set of user-defined robustness metrics. The package is based on the recently proposed Recurrent Equilibrium Network (REN) and Lipschitz-Bounded Deep Network (LBDN) model classes, and is designed to interface directly with Julia's most widely-used machine learning package, Flux.jl. We discuss the theory behind our model parameterization, give an overview of the package, and provide a tutorial demonstrating its use in image classification, reinforcement learning, and nonlinear state-observer design.
title RobustNeuralNetworks.jl: a Package for Machine Learning and Data-Driven Control with Certified Robustness
topic Machine Learning
Systems and Control
url https://arxiv.org/abs/2306.12612