Model Monitoring Best Practices
Whitepaper
Lessons from the field on model monitoring best practices
A growing number of decisions and critical business processes rely on models produced with machine learning and other statistical techniques.
For a variety of reasons, the inputs and outputs from these models can “drift” over time, and produce unexpected behavior and a decrease in predictive accuracy. Unfortunately, this drift often goes unrecognized because of inadequate tools or internal processes, leading to severe financial loss or a degraded customer experience.
Download this paper today to learn:
- The types of data drift that impact machine learning models;
- How to identify models that are degrading;
- Five best practices for monitoring models in production; and
- Recommended next steps for correcting model drift.
This paper represents best practices Domino has learned from 5+ years of working with data science leaders at companies such as Allstate, Bayer, Dell and Moody’s Analytics.
