Xin Xu
ABSTRACT
With the rapid development of deep learning technology, the application based on deep learning shows explosive growth, especially in the field of computer vision. However, some characteristics of the deep learning algorithm make it face many unexpected threats in actual use. It has become an urgent problem to ensure the reliability of the algorithm. In this paper, we propose a deep learning algorithm reliability evaluation system and corresponding evaluation indicators from the three aspects of data, model, and operating framework, then describe reliability evaluation methods for each specific indicator.
- F. Schroff, D. Kalenichenko, J. Philbin. 2018. FaceNet: A unified embedding for face recognition and clustering. In Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR 2015), 815--823.Google Scholar
- W. Zhu, Y. S. Vang, Y. Huang, and X. Xie. 2018. Deepem: Deep 3d convnets with em for weakly supervised pulmonary nodule detection. In International Conference on Medical Image Computing and Computer-Assisted Intervention, 812--820.Google Scholar
- P. Li, X. Chen, and S. Shen, 2019. Stereo r-cnn based 3d object detection for autonomous driving. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2019), 7644--7652.Google Scholar
- M. L. Iuzzolino, M. E. Walker, and D. Szafir, 2018. Virtual-to-Real-World Transfer Learning for Robots on Wilderness Trails. In 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2018), 576--582.Google Scholar
- Y. He, G. Meng, K. Chen, X. Hu, and J. He. 2019. Towards privacy and security of deep learning systems: a survey. arXiv preprint arXiv:1911.12562.Google Scholar
- Y. LeCun, Y. Bengio, and G. Hinton. 2015. Deep learning. nature, 521(7553), 436--444.Google Scholar
- P. Santhanam, E. Farchi, and V. Pankratius. 2019. Engineering Reliable Deep Learning Systems. arXiv preprint arXiv:1910.12582.Google Scholar
- F. F. Li, R. Krishna, and D. F. Xu, http://cs231n.stanford.edu/slides/2020/lecture_13.pdfGoogle Scholar
- I. J. Goodfellow, J. Shlens, and C. Szegedy. 2015. Explaining and harnessing adversarial examples. international conference on learning representations (ICLR 2015). arXiv preprint arXiv:1412.6572, 1.Google Scholar
- S. M. Moosavi-Dezfooli, A. Fawzi, and P. Frossard. 2016. Deepfool: a simple and accurate method to fool deep neural networks. In Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR 2016), 2574--2582.Google Scholar
- N. Carlini and D. Wagner. 2017. Towards evaluating the robustness of neural networks. In 2017 ieee symposium on security and privacy (sp), 39--57.Google Scholar
- A. Madry, A. Makelov, L. Schmidt, D. Tsipras, and A. Vladu. 2018. Towards Deep Learning Models Resistant to Adversarial Attacks. international conference on learning representations (ICLR 2018).Google Scholar
- P. Molchanov, S. Tyree, T. Karras, T. Aila, and J. Kautz. 2017. Pruning Convolutional Neural Networks for Resource Efficient Inference. international conference on learning representations (ICLR 2017).Google Scholar
- M. Lecuyer, V. Atlidakis, R Geambasu, D. Hsu, and S. Jana. 2019. Certified robustness to adversarial examples with differential privacy. In 2019 IEEE Symposium on Security and Privacy (SP), 656--672.Google Scholar
Index Terms
- Reliability Evaluation Methods of Deep Learning Algorithm in Computer Vision
Recommendations
A Systematic Mapping Study of Computer Vision Approaches based on Deep Learning and Neural Network
BCI '17: Proceedings of the 8th Balkan Conference in InformaticsDeep architectures with convolution structure have been found highly effective and been commonly used in computer vision and image recognition. With the introduction of Graphics Processing Unit (GPU) for general purpose issues, there has been an ...
Deep reinforcement learning in computer vision: a comprehensive survey
AbstractDeep reinforcement learning augments the reinforcement learning framework and utilizes the powerful representation of deep neural networks. Recent works have demonstrated the remarkable successes of deep reinforcement learning in various domains ...
Recent Progress in Deep Reinforcement Learning for Computer Vision and NLP
RFIW '17: Proceedings of the 2017 Workshop on Recognizing Families In the WildDeep reinforcement learning is considered as a way of building autonomous system with a higher level understanding of the world and would revolutionize the field of AI. Recently, some researchers have made many progresses such as learning to play video ...
Comments