AI Quality Workshop

ML Explainability - How do you explain model predictions? What are best practices for local and global explanations, and conceptual soundness assessment? Which methods are most appropriate for various use cases?
Accuracy and Performance Debugging - How can you systematically analyze model accuracy? How can you rapidly identify model errors to drive improvements?
Model Drift - Does your production model need to be refreshed? How do you measure model score and data drift on an ongoing basis? How can you understand the root causes of drift, and debug your models in a directed way?
Fairness - How can you ensure that your models are set up to be fair and compliant, and remain fair over time? What are best practices for choosing fairness metrics, understanding root causes and mitigating fairness gaps, leveraging humans in the loop?
NLP Model Quality - What are the specific challenges of NLP model quality? What are best practices for explainability, debugging and improving the performance of NLP models?