PV1 BDD4BNNencodes a binarized neural network and input region into a BDD

Framework for quantitative analysis of BNNs (binarized neural networks).

Application domain/field

Expected input

Expected output

BDD that encodes the input-output relation of the binarized neural network in the region R.

Internals

It encodes a BNN and the associated input region into BDDs. Uses CUDD

Comments

The tool is called a "prototype tool" in the CAV '21 paper.
BDD Library

Links

Cannot find a repository or download link

Related papers

BDD4BNN: A BDD-Based Quantitative Analysis Framework for Binarized Neural Networks (CAV '21)

Last publication date

15 July 2021

ProVerB specific



ProVerB is a part of SLEBoK. Last updated: February 2023.