- 专利标题: Fraud score manipulation in self-defense of adversarial artificial intelligence learning
-
申请号: US15590921申请日: 2017-05-09
-
公开(公告)号: US11100506B2公开(公告)日: 2021-08-24
- 发明人: Scott Michael Zoldi , Qing Liu
- 申请人: FAIR ISAAC CORPORATION
- 申请人地址: US MN Roseville
- 专利权人: FAIR ISAAC CORPORATION
- 当前专利权人: FAIR ISAAC CORPORATION
- 当前专利权人地址: US MN Roseville
- 代理机构: Mintz, Levin, Cohn, Ferris, Glovsky and Popeo, P.C
- 代理商 Michael Van Loy; Paul Brockland
- 主分类号: G06Q20/40
- IPC分类号: G06Q20/40 ; G06N3/08 ; G06N20/00 ; G06N3/04
摘要:
A system and method for programmatically revealing misleading confidence values in Fraud Score is presented to protect artificial intelligence models from adversarial neural networks. The method is used to reduce an adversarial learning neural network model effectiveness. With the score manipulation implemented, the adversary models are shown to systematically become less successful in predicting the true behavior of the Fraud detection artificial intelligence model and what it will flag as fraudulent transactions, thus reducing the true fraud dollars penetrated or taken by adversaries.