Jan Hendrik Metzen
15Patents
2h-index
18Co-inventors
43Inventor score
Filing activity: Jan 31, 2018 → Aug 24, 2022
Most-cited inventions
| Patent | Title | Area | Cited by | Status |
|---|---|---|---|---|
| US10824122B2 | Method and device for improving the robustness with respect to “adversarial examples” | Physics | 35 | Active |
| US11676025B2 | Method, apparatus and computer program for generating robust automatic learning systems and testing trained automatic learning systems | Physics | 2 | Active |
| US11947625B2 | Data-based updating of the training of classifier networks | Physics | 0 | Active |
| US12243283B2 | Device and method for determining a semantic segmentation and/or an instance segmentation of an image | Physics | 0 | Active |
| US10331091B2 | Method and device for improving the robustness with respect to “adversarial examples” | Physics | 0 | Active |
| US12086214B2 | Method and device for creating a machine learning system | Physics | 0 | Active |
| US11599827B2 | Method and apparatus for improving the robustness of a machine learning system | Physics | 0 | Active |
| US11055632B2 | Method and device for improving the robustness against “adversarial examples” | Physics | 0 | Active |
| US12205349B2 | System and method for improving robustness of pretrained systems in deep neural networks utilizing randomization and sample rejection | Physics | 0 | Active |
| US12340572B2 | Device and method to adapt a pretrained machine learning system to target data that has different distribution than the training data without the necessity of human annotations on target data | Physics | 0 | Active |
| US11727277B2 | Method and apparatus for automatically producing an artificial neural network | Physics | 0 | Active |
| US11500998B2 | Measuring the vulnerability of AI modules to spoofing attempts | Physics | 0 | Active |
| US12204653B2 | System and method, in particular computer implemented method for determining a perturbation for attacking and/or validating an object tracker | Physics | 0 | Active |
| US11138467B2 | Method, device, product, and computer program for operating a technical system | Physics | 0 | Active |
| US12014280B2 | Measuring the sensitivity of neural network image classifiers against adversarial attacks | Physics | 0 | Active |
Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.