Machine Learning IP Protection

Submitted by grigby1 on Thu, 07/30/2020 - 2:05pm

Title	Machine Learning IP Protection
Publication Type	Conference Paper
Year of Publication	2018
Authors	Cammarota, Rosario, Banerjee, Indranil, Rosenberg, Ofer
Conference Name	2018 IEEE/ACM International Conference on Computer-Aided Design (ICCAD)
Date Published	November 2018
Publisher	ACM
Keywords	application domains, composability, computer security, Deep Learning, Human Behavior, human factors, industrial property, Intellectual Properties (IP), intellectual property, intellectual property security, ip protection, learning (artificial intelligence), machine learning, Machine Learning IP protection, Metrics, policy-based governance, proprietary machine learning models, pubcrawl, resilience, Resiliency, security of data, system security architecture mechanisms, Trusted Computing
Abstract	Machine learning, specifically deep learning is becoming a key technology component in application domains such as identity management, finance, automotive, and healthcare, to name a few. Proprietary machine learning models - Machine Learning IP - are developed and deployed at the network edge, end devices and in the cloud, to maximize user experience. With the proliferation of applications embedding Machine Learning IPs, machine learning models and hyper-parameters become attractive to attackers, and require protection. Major players in the semiconductor industry provide mechanisms on device to protect the IP at rest and during execution from being copied, altered, reverse engineered, and abused by attackers. In this work we explore system security architecture mechanisms and their applications to Machine Learning IP protection.
URL	https://dl.acm.org/doi/10.1145/3240765.3270589
DOI	10.1145/3240765.3270589
Citation Key	cammarota_machine_2018

Groups:

Science of Security VO