Concurrent Technologies Corporation (CTC) is performing research and development into the emerging fields of ML interpretability & assurance to better understand the trade-offs between model utility, privacy, security, and interpretability. In this research, we leverage novel techniques to understand model processing and how to contextualize model output with additional information. This repo demonstrates some of the tools and techniques that our analysis use to better interpret modern deep neural networks.
More information on CTC's AI & ML capabilities can be found here.