Adversarial training is not safe for mobile robots
bdtechtalks.substack.com
Adversarial attacks have become a common concern in deep learning. Machine learning engineers train their deep neural networks on adversarial examples to make them less sensitive to adversarial perturbations. But the same process, called “adversarial training,” can cause unwanted side effects when applied to robotics. According to research by scientists at the Institute of Science and Technology Austria, the Massachusetts Institute of Technology, and Technische Universitat Wien, Austria, adversarial training reduces the safety of neural networks and creates new error profiles in robotics applications.
Adversarial training is not safe for mobile robots
Adversarial training is not safe for mobile…
Adversarial training is not safe for mobile robots
Adversarial attacks have become a common concern in deep learning. Machine learning engineers train their deep neural networks on adversarial examples to make them less sensitive to adversarial perturbations. But the same process, called “adversarial training,” can cause unwanted side effects when applied to robotics. According to research by scientists at the Institute of Science and Technology Austria, the Massachusetts Institute of Technology, and Technische Universitat Wien, Austria, adversarial training reduces the safety of neural networks and creates new error profiles in robotics applications.