TruEra Research: Explainable ML A core research direction for TruEra is studying how to robustly explain models in order to understand, introspect, and trust them. TruEra solutions are based on years of explainability research conducted at Carnegie Mellon University. We continue to view explainability as the backbone for trust in ML systems.
TruEra Diagnostics is an AI Quality solution that helps data scientists to analyze and optimize machine learning model performance, explain model function, and minimize algorithmic bias. With TruEra Diagnostics, data scientists can create high quality models, faster, as well as demonstrate to key stakeholders that their models are ready for production and meet customer or regulatory requirements. TruEra Diagnostics works across both custom models and models created with the most popular model development platforms, such as Data Robot, H20.ai, and Dataiku. It also works across a variety of model serving providers, and fits easily into the AI stack.
TruEra Monitoring helps you easily track and troubleshoot machine learning model performance. With unique explainability and model quality analytics, TruEra Monitoring goes beyond basic observability solutions by enabling faster root cause analysis and action. This saves ML ops and data scientist time, improves governance, and provides a more effective feedback loop to improve both models and business outcomes.