Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks
Engineers explore architecture and security mechanisms for Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks.
Summary
No summary yet — watch the video above, or open it on YouTube.
Related talks
IEEE Symposium on Security and Privacy·15 min
ShadowNet: A Secure and Efficient On-device Model Inference System for Convolutional Neural Networks
Security researchers examine technical aspects of ShadowNet: A Secure and Efficient On-device Model Inference System for Convolutional Neural Networks.
neural-networksnetworksshadownet+3
IEEE Symposium on Security and Privacy·15 min
Security researchers examine technical aspects of SoK: Certified Robustness for Deep Neural Networks.
neural-networksrobustnessnetworks
IEEE Symposium on Security and Privacy·16 min
Industry experts analyze emerging threats surrounding LinkTeller: Recovering Private Edges from Graph Neural Networks via Influence Analysis.
neural-networksgraphnetworks