Utilizing data provenance to defend against data poisoning attacks
The present invention discloses a secure ML pipeline to improve the robustness of ML
models against poisoning attacks and utilizing data provenance as a tool. Two components …
models against poisoning attacks and utilizing data provenance as a tool. Two components …
Creating an effective product using an attribute solver
AD Yates - US Patent 11,790,030, 2023 - Google Patents
Disclosed here is a system that can obtain attributes of an advertisement, where an attribute
has a continuous value, and a range of acceptable values is uncertain. The system can …
has a continuous value, and a range of acceptable values is uncertain. The system can …
Techniques for understanding how trained neural networks operate
AC Öztireli, M Gross, M Ancona - US Patent 11,568,212, 2023 - Google Patents
2019-08-08 Assigned to ETH Zürich, THE WALT DISNEY COMPANY (SWITZERLAND)
GMBH reassignment ETH Zürich ASSIGNMENT OF ASSIGNORS INTEREST (SEE …
GMBH reassignment ETH Zürich ASSIGNMENT OF ASSIGNORS INTEREST (SEE …