NVIDIA Research: Auditing AI Models for Verified Deployment under Semantic Specifications

Originally published at: https://developer.nvidia.com/blog/nvidia-research-auditing-ai-models-for-verified-deployment-under-semantic-specifications/

○ Quality assurance and audits are necessary for deep learning models. Current AI models require large data sets for training or a designed reward function that must be optimized. Algorithmically, AI is prone to optimizing behaviors that were not intended by the human designer. To help combat this, the AuditAI framework was developed to help audit these problems, which increases safety and ethical use of deep learning models during deployment.

Do deep learning models need auditing? Find out about AuditAI and see how you can benefit from QA for your AI model.

What is the link for additional information?
“For more information, see Auditing AI models for Verified Deployment under Semantic Specifications [LINK].”

Good catch, @medgar! The paper is still under review, so I’ll add that link when the paper’s available online.

The paper with technical details is now available:

@medgar and @animeshg, the post is now updated with the link. Thanks!