Model transparency for predictive models
Transparent artificial intelligence is becoming an important requirement for many businesses. In risk management, decisions need to be explainable, and opaque predictive models are not allowed. In marketing, the policy for the transparency of models might be less strict and allow for the use of opaque models.
Each predictive model type that comes with Pega® Platform is assigned a transparency score by default. For example, a decision tree has a high transparency score, whereas a neural network model has a low transparency score. By default, the transparency threshold is set to 1 and all model types are allowed in all business issues. Lead data scientists can modify transparency thresholds for different business issues. For example, they can increase the threshold for risk management to indicate that opaque models are non-compliant in that area.
Model Transparency Policy landing page
Model transparency policy in Pega Platform helps you indicate compliant and non-compliant predictive models.
Compliant and non-compliant models
When you develop models in the Analytics Center portal, you can check the transparency policy on the portal at any time.
Viewing a model transparency policy in the Analytics Center portal
For more information, see Configuring the model transparency policy for predictive models. (LINK to -> dsm\tasks\dsm-configuring-models-transparency-tsk.htm)
Published September 20, 2017 — Updated August 29, 2018