arXiv Open Access 2024

Data-driven Verification of DNNs for Object Recognition

Clemens Otte Yinchong Yang Danny Benlin Oswan

Lihat Sumber

Abstrak

The paper proposes a new testing approach for Deep Neural Networks (DNN) using gradient-free optimization to find perturbation chains that successfully falsify the tested DNN, going beyond existing grid-based or combinatorial testing. Applying it to an image segmentation task of detecting railway tracks in images, we demonstrate that the approach can successfully identify weaknesses of the tested DNN regarding particular combinations of common perturbations (e.g., rain, fog, blur, noise) on specific clusters of test images.

Topik & Kata Kunci

cs.CV

Penulis (3)

Clemens Otte

Yinchong Yang

Danny Benlin Oswan

Format Sitasi

APA MLA BibTeX

Otte, C., Yang, Y., Oswan, D.B. (2024). Data-driven Verification of DNNs for Object Recognition. https://arxiv.org/abs/2408.00783

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2024
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓