RobustNeuralNetworks.jl: a Package for Machine Learning and Data-Driven Control with Certified Robustness
This package addresses robustness issues in neural networks for machine learning and control applications, but it is incremental as it builds on existing model classes like REN and LBDN.
The authors tackled the problem of neural networks being sensitive to small input perturbations by introducing RobustNeuralNetworks.jl, a Julia package that constructs models to naturally satisfy user-defined robustness metrics, with applications demonstrated in image classification, reinforcement learning, and nonlinear state-observer design.
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.