eduzhai > Applied Sciences > Engineering >

Green Lighting ML Confidentiality Integrity and Availability of Machine Learning Systems in Deployment

  • king
  • (0) Download
  • 20210505
  • Save

... pages left unread,continue reading

Document pages: 3 pages

Abstract: Security and ethics are both core to ensuring that a machine learning systemcan be trusted. In production machine learning, there is generally a hand-offfrom those who build a model to those who deploy a model. In this hand-off, theengineers responsible for model deployment are often not privy to the detailsof the model and thus, the potential vulnerabilities associated with its usage,exposure, or compromise. Techniques such as model theft, model inversion, ormodel misuse may not be considered in model deployment, and so it is incumbentupon data scientists and machine learning engineers to understand thesepotential risks so they can communicate them to the engineers deploying andhosting their models. This is an open problem in the machine learning communityand in order to help alleviate this issue, automated systems for validatingprivacy and security of models need to be developed, which will help to lowerthe burden of implementing these hand-offs and increasing the ubiquity of theiradoption.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×