Tiny alterations in training data can introduce “backdoors” into machine learning models
Tiny alterations in training data can introduce “backdoors” into machine learning models
In TrojDRL: Trojan Attacks on Deep Reinforcement Learning Agents, a group of Boston University researchers demonstrate an attack on machine learning systems trained with “reinforcement learniâ¦
Tiny alterations in training data can introduce “backdoors” into machine learning models
Hugh Taylor
Hugh Taylor is a Certified Information Security Manager (CISM). In addition to editing Journal of Cyber Policy, he writes about cybersecurity, compliance and enterprise technology for such clients as Microsoft, IBM, SAP, HPE, Oracle, Google and Advanced Micro Devices. Prior to launching his freelance writing career, he served in executive roles at Microsoft, IBM and several venture-backed technology startups.