alpha-beta-CROWN: An Efficient, Scalable and GPU Accelerated Neural Network Verifier (winner of VNN-COMP 2021, 2022, 2023, and 2024)
-
Updated
Nov 25, 2024 - Python
alpha-beta-CROWN: An Efficient, Scalable and GPU Accelerated Neural Network Verifier (winner of VNN-COMP 2021, 2022, 2023, and 2024)
Neural Network Verification Software Tool
Lyapunov-stable Neural Control for State and Output Feedback
DIG is a numerical invariant generation tool. It infers program invariants or properties over (i) program execution traces or (ii) program source code. DIG supports many forms of numerical invariants, including nonlinear equalities, octagonal and interval properties, min/max-plus relations, and congruence relations.
NCubeV - The Nonlinear Neural Network Verifier
Uses the simplex to propose a tighter boundary for the l1 perturbation of the convex activation function network, improving the effect of the CROWN algorithm.
Verification of neural networks based on input splitting and forward propagation of symbolic intervals with fresh variables.
This repository contains a collection of residual benchmarks from VNNCOMP 2022-2024. It is designed to offer a more organized version of the existing benchmarks, making it easier to test new software. We recommend cloning the 'benchmarks_vnncomp' repository, which includes this repository as a submodule.
This repository contains a collection of fully connected benchmarks from VNNCOMP 2022-2024. It is designed to offer a more organized version of the existing benchmarks, making it easier to test new software. We recommend cloning the 'benchmarks_vnncomp' repository, which includes this repository as a submodule.
This repository contains a collection of convolutional benchmarks from VNNCOMP 2022-2024. It is designed to offer a more organized version of the existing benchmarks, making it easier to test new software. We recommend cloning the 'benchmarks_vnncomp' repository, which includes this repository as a submodule.
Verification of Neural Network Control Systems in Continuous Time
A reorganized collection of benchmarks from VNNCOMP since 2022, divided into three categories: fully connected, convolutional, and residual networks. Each category is available as a submodule, allowing you to download individual categories or all of them at once.
We find the largest region where a given property of a pre-trained neural network is verified.
Add a description, image, and links to the neural-network-verification topic page so that developers can more easily learn about it.
To associate your repository with the neural-network-verification topic, visit your repo's landing page and select "manage topics."