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

Traffic signs play a critical role in road safety and traffic management forautonomous driving systems. Accurate traffic sign classification is essentialbut challenging due to real-world complexities like adversarial examples andocclusions. To address these issues, binary neural networks offer promise inconstructing classifiers suitable for resource-constrained devices.  In our previous work, we proposed high-accuracy BNN models for traffic signrecognition, focusing on compact size for limited computation and energyresources. To evaluate their local robustness, this paper introduces a set ofbenchmark problems featuring layers that challenge state-of-the-artverification tools. These layers include binarized convolutions, max pooling,batch normalization, fully connected. The difficulty of the verificationproblem is given by the high number of network parameters (905k - 1.7 M), ofthe input dimension (2.7k-12k), and of the number of regions (43) as well bythe fact that the neural networks are not sparse.  The proposed BNN models and local robustness properties can be checked athttps://github.com/ChristopherBrix/vnncomp2023_benchmarks/tree/main/benchmarks/traffic_signs_recognition.  The results of the 4th International Verification of Neural NetworksCompetition (VNN-COMP'23) revealed the fact that 4, out of 7, solvers canhandle many of our benchmarks randomly selected (minimum is 6, maximum is 36,out of 45). Surprisingly, tools output also wrong results or missingcounterexample (ranging from 1 to 4). Currently, our focus lies in exploringthe possibility of achieving a greater count of solved instances by extendingthe allotted time (previously set at 8 minutes). Furthermore, we are intriguedby the reasons behind the erroneous outcomes provided by the tools for certainbenchmarks.

Benchmarking Local Robustness of High-Accuracy Binary Neural Networks for Enhanced Traffic Sign Recognition