Patent Leaderboard
USPTO Patent Rankings Data through Dec 31, 2025
Home › Using gradients to detect backdoors in neural networks

Using gradients to detect backdoors in neural networks

US Patent 11132444 · Granted Sep 28, 2021

Estimated economic value: $6,479,000 ⓘ

Assignee

Inventors

View full patent text on Google Patents →

More Patents by These Inventors

Graph exploration framework for adversarial example generation2025
Graph exploration framework for adversarial example generation2025
Equalizer for removing inter symbol interference of data signal by increasing pu2025
Adaptive robustness certification against adversarial examples2025
Adaptive robustness certification against adversarial examples2025
Methods of fabricating semiconductor devices2025
Federated learning with dataset sketch commitment based malicious participant id2024
Dynamic gradient deception against adversarial examples in machine learning mode2024
Dynamic gradient deception against adversarial examples in machine learning mode2024
Detecting adversary attacks on a deep neural network (DNN)2024
Detecting adversary attacks on a deep neural network (DNN)2024
Semiconductor device including delay compensation circuit2024