A toolkit for enforcing logical specifications on neural networks
See the codeREADME
Vehicle
Vehicle is a system for embedding logical specifications into neural networks. At its heart is the Vehicle specification language, a high-level, functional language for writing mathematically-precise specifications for your networks.
These specifications can then automatically be compiled down to loss functions to be used when training your network. After training, the same specification can be compiled down to queries for low-level neural network solvers which either prove that the specification holds or produce a counter-example. Such a proof is far better than simply testing, as you can prove that the specification holds for all inputs. Verified specifications can also be exported to interactive theorem provers (ITPs) such as Rocq, Isabelle/HOL and Agda. This in turn allows for the formal verification of larger software systems that use neural networks as subcomponents. The generated ITP code is tightly linked to the actual deployed network, so changes to the network will result in errors when checking the larger proof.
Documentation
Examples
Each of the following examples comes with an explanatory README file:
-
ACAS Xu - The complete specification of the ACAS Xu collision avoidance system from the Reluplex paper in a single file.
-
Car controller - A neural network controller that is formally proven to always keep a simple model of a car on the road in the face of noisy sensor data and an unpredictable cross-wind.
-
MNIST robustness - A classifier for the MNIST dataset that is proven to be robust around the images in the dataset.
In addition to the above, further examples of specifications can be found in the test suite and the corresponding output of the Vehicle compiler can be found here.
Support
If you are interested in adding support for a particular format or tool then open an issue on the Issue Tracker to discuss it with us.
Neural network formats
Solver backends
- VNNLIB 2.0 compatible solvers
- Marabou
Interactive Theorem Prover backends
Dataset formats
Related papers
Contributors
Languages
Haskell
85.4%
Python
9.2%
VCL
2.5%
Shell
1.1%
A toolkit for enforcing logical specifications on neural networks
See the codeREADME
Vehicle
Vehicle is a system for embedding logical specifications into neural networks. At its heart is the Vehicle specification language, a high-level, functional language for writing mathematically-precise specifications for your networks.
These specifications can then automatically be compiled down to loss functions to be used when training your network. After training, the same specification can be compiled down to queries for low-level neural network solvers which either prove that the specification holds or produce a counter-example. Such a proof is far better than simply testing, as you can prove that the specification holds for all inputs. Verified specifications can also be exported to interactive theorem provers (ITPs) such as Rocq, Isabelle/HOL and Agda. This in turn allows for the formal verification of larger software systems that use neural networks as subcomponents. The generated ITP code is tightly linked to the actual deployed network, so changes to the network will result in errors when checking the larger proof.
Documentation
Examples
Each of the following examples comes with an explanatory README file:
-
ACAS Xu - The complete specification of the ACAS Xu collision avoidance system from the Reluplex paper in a single file.
-
Car controller - A neural network controller that is formally proven to always keep a simple model of a car on the road in the face of noisy sensor data and an unpredictable cross-wind.
-
MNIST robustness - A classifier for the MNIST dataset that is proven to be robust around the images in the dataset.
In addition to the above, further examples of specifications can be found in the test suite and the corresponding output of the Vehicle compiler can be found here.
Support
If you are interested in adding support for a particular format or tool then open an issue on the Issue Tracker to discuss it with us.
Neural network formats
Solver backends
- VNNLIB 2.0 compatible solvers
- Marabou
Interactive Theorem Prover backends
Dataset formats
Related papers
Contributors
Languages
Haskell
85.4%
Python
9.2%
VCL
2.5%
Shell
1.1%