Data-Driven Certification of Neural Networks With Random Input Noise-Reference-Cited by-同舟云学术

Data-Driven Certification of Neural Networks With Random Input Noise

Published:2023-03 Issue:1 Volume:10 Page:249-260
ISSN:2325-5870
Container-title:IEEE Transactions on Control of Network Systems
language:
Short-container-title:IEEE Trans. Control Netw. Syst.

Author:

Anderson Brendon G.¹^ORCID,Sojoudi Somayeh²^ORCID

Affiliation:

1. Department of Mechanical Engineering, University of California, Berkeley, Berkeley, CA, USA

2. Department of Electrical Engineering and Computers Sciences and the Department of Mechanical Engineering, University of California, Berkeley, Berkeley, CA, USA

Funder

NSF

ONR YIP

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

Control and Optimization,Computer Networks and Communications,Signal Processing,Control and Systems Engineering

Link

https://ieeexplore.ieee.org/ielam/6509490/10075640/9857969-aam.pdf

Reference38 articles.

1. A statistical approach to assessing neural network robustness;webb;Proc Int Conf Learn Represent,0

2. A General Scenario Theory for Nonconvex Optimization and Decision Making

3. Tightened Convex Relaxations for Neural Network Robustness Certification

4. A Sample Approximation Approach for Optimization with Probabilistic Constraints

Cited by 2 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. Improving the Accuracy-Robustness Trade-Off of Classifiers via Adaptive Smoothing;SIAM Journal on Mathematics of Data Science;2024-08-14

2. Efficient Global Optimization of Two-Layer ReLU Networks: Quadratic-Time Algorithms and Adversarial Training;SIAM Journal on Mathematics of Data Science;2023-06-01