Affiliation:
1. Laboratoire Central de Recherches, Thomson-CSF 91404 Orsay (Cedex), France
Abstract
In this paper, we study the robustness of multilayer networks versus the destruction of neurons. We show that the classical backpropagation algorithm does not lead to optimal robustness and we propose a modified algorithm that improves this capability.
Subject
Cognitive Neuroscience,Arts and Humanities (miscellaneous)
Cited by
31 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Towards Robust Deep Neural Networks Against Design-Time and Run-Time Failures;2023 IEEE International Test Conference (ITC);2023-10-07
2. Artificial Intelligence Techniques within a Technological Process of Manufacturing Automated Control Systems;2022 International Conference on Quality Management, Transport and Information Security, Information Technologies (IT&QM&IS);2022-09-26
3. Detecting Functional Safety Violations in Online AI Accelerators;2022 IEEE 28th International Symposium on On-Line Testing and Robust System Design (IOLTS);2022-09-12
4. Robustness assessment and improvement of a neural network for blood oxygen pressure estimation;2022 IEEE Conference on Software Testing, Verification and Validation (ICST);2022-04
5. Toward Functional Safety of Systolic Array-Based Deep Learning Hardware Accelerators;IEEE Transactions on Very Large Scale Integration (VLSI) Systems;2021-03