Partial Dependence (PDPs) and Individual Conditional Expectation (ICE) Plots | Intuition and Math

Both Partial Dependence (PDPs) and Individual Conditional Expectation (ICE) Plots are used to understand and explain machine learning models. PDPs can tell us if a relationship between a model feature and target variable is linear, non-linear or if there is no relationship. Similarly, ICE plots are used to visualise interactions. Now, at first glance, these plots may look complicated. But you will see, they are actually constructed in a fairly intuitive way.

In this video, we will:
- Take you step-by-step through how PDPs and ICE plots are created.
- Discuss what insight the explainable AI methods can give
- And we will end by explaining the mathematics behind PDPs.

🚀 Free Course 🚀

🚀 Companion article with link to code (no-paywall link): 🚀

🚀 Useful playlists 🚀

🚀 Get in touch 🚀

🚀 Chapters 🚀
00:00 Introduction
01:31 Understanding PDPs
04:20 Visualising relationships with PDPs
06:58 Understanding ICE Plots
06:26 The maths

[makefly]

Hi I have a question at 5:45, wanna know based on which pattern of the plot you said the "km_driven" is less equally distributed and skewed to the left? 😄

[IsmaelSilva-poxb]

Writing a subsection on ICE on my master dissertation

[TheCsePower]

Would be nice if the pdp had some kind of confidence interval that varied with the feature value.