Decisions have proven to be a core component of an enterprise digital transformation. It is important to note that digital decisions – along with machine learning and mathematical optimization – are keys to realizing artificial intelligence. The vast amount of data available – particularly in mortgage lending – makes this an exciting area of exploration. However, it must be approached judiciously.

In this presentation we present the concept of a decision framework for sensibly enabling machine learning in the enterprise in three key areas:
1. Construction of a permissible use and compliance process that ensures data and predictive models created from that data are properly vetted across numerous constraints.
2. Development of a human and machine hybrid model of boundary rules to assure proper use of data and predictive models – an “enforcement level” of decisions.
3. Establishing a foundation for explainable AI using decisions. 
This framework will be discussed using numerous examples relevant to financial services and lending.

Learning Objectives:

• Application of Decisions to Machine Learning
• Approaches to Explainable AI
• Use of Augmented AI