A new perspective on Shapley values, part II: The Naïve Shapley method by Edden Gerber

Why should you read this post?
1. For insight into Shapley values and the SHAP tool.
Most other sources on these topics are explanations based on existing primary sources (e.g. academic papers and the SHAP documentation). This post is an attempt to gain some understanding through an empirical approach.
2. To learn about an alternative approach to computing Shapley values, that under some (limited) circumstances may be preferable to SHAP.
If you are unfamiliar with Shaply values or SHAP, or want a short recap of how the SHAP explainers work, check out the previous post. In a hurry? The author has emphasized the key sentences in bold to assist your speed-reading.

Link: Post

Navigational hashtags: #armknowledgesharing #armsites
General hashtags: #shap #shapley #interpretation #ml #python

@data_science_weekly