English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Many approaches for verifying input-output properties of neural networks havebeen proposed recently. However, existing algorithms do not scale well to largenetworks. Recent work in the field of model compression studied binarizedneural networks (BNNs), whose parameters and activations are binary. BNNs tendto exhibit a slight decrease in performance compared to their full-precisioncounterparts, but they can be easier to verify. This paper proposes a simplemixed integer programming formulation for BNN verification that leveragesnetwork structure. We demonstrate our approach by verifying properties of BNNstrained on the MNIST dataset and an aircraft collision avoidance controller. Wecompare the runtime of our approach against state-of-the-art verificationalgorithms for full-precision neural networks. The results suggest that thedifficulty of training BNNs might be worth the reduction in runtime achieved byour verification algorithm.

A Mixed Integer Programming Approach for Verifying Properties of Binarized Neural Networks