Watch this Ask the Expert session to learn techniques for interpreting machine learning models and how to use them in SAS Viya.
Watch the Webinar
You will learn:
-
What are black box models?
-
Why interpretability matters.
-
When to distinguish between local and global explanations.
The questions from the Q&A segment held at the end of the webinar are listed below and the slides from the webinar are attached.
Q&A
What is the difference between Individual conditional expectation curve and linear regression curve, does same task of prediction?
No. In ICE, given a chosen observation, we analyze the behavior of a variable, keeping the values of the other variables constant and changing the value of the variable under analysis. For each value reinforced by the variable under analysis, the response is obtained using the model. In this way, we can analyze the behavior of the model, for a given instance when the value of a variable is changed.
In LIME, synthetic samples are created from a specific instance based on small variations of its attributes. The responses of each observation of this new synthetic data set are obtained after the same presentations to the model under analysis. Then, a linear regression model is obtained from the data set formed by the synthetic data with their respective targets.
Recommended Resources
Demystifying the Black Box: How Explainable AI is Building Trust in AI Systems
Explainable Artificial Intelligence for Anti-Money Laundering;
What is Explainable AI?;
What is Explainable AI (XAI)? | IBM
Tech Dispatch | Explainable Artificial Intelligence
Please see additional resources in the attached slide deck.
Want more tips? Be sure to subscribe to the Ask the Expert board to receive follow up Q&A, slides and recordings from other SAS Ask the Expert webinars.
