Explainable monitoring – governance
In this section, we will implement the governance mechanisms that we learned about previously in Chapter 11, Key Principles of Monitoring Your ML System, for the business use case we have been working on. We will delve into three of the components of governing an ML system, as shown in the following diagram:
Figure 13.3 – Components of governing your ML system
The effectiveness of ML systems results from how they are governed to maximize business value. To have end-to-end trackability and comply with legislation, system governance requires quality assurance and monitoring, model auditing, and reporting. We can regulate and rule ML systems by monitoring and analyzing model outputs. Smart warnings and behavior guide governance to optimize business value. Let's look at how the ML system's governance is orchestrated by warnings and behavior, model quality assurance and control, model auditing, and reports...
