Machine Learning Explainability Training

Get a toolbox of interpretability techniques you can use in your daily work, understanding when and how you can and should use them.

Start by applying techniques based on implementations from popular packages and what their drawbacks are to prevent investigating blackbox models with blackbox interpretability techniques.

Requirements

You should be proficient with scikit-learn:

• Pipelines
• Column transformers
• Linear models, and more complex models (e.g. random forests and gradient boosting).

What you'll learn

• Understand the use cases for model explainability: debug, right to explanation, etc.
• Understand when model explainability is not enough (e.g. causality and fairness).
• Categorize the methods covered between sensitivity/impact and global/local.
• Apply the methods with the provided packages.
• Explain the inner workings of all methods.
• Articulate the downsides of each method.
• Evaluate whether a method is appropriate for the business use case.

Using scikit-learn, dalex, and shap, we will look at:

• Addictive Attributions.
• Partial Dependence Plots.
• Individual Conditional Expectation Curves.
• Permutation Importance.
• Shapley Values.

Our Data Science learning journey

After following the Machine Learning Explainability training course, you can also deepen your knowledge as a Data Science. Our courses combine just the right blend of theory with practical exercises.

Check out the Data Science learning journey at Xebia Data.

What else should I know?

• Available online instructor-led, classroom, and in-company. This training can be delivered both in-person and online. When hosting the in-person training, we provide lunch, snacks, and drinks to the participants. Accordingly, there is a discount for virtual training courses.
• Don't worry, you don't have to install any software on your laptop.
• This course is brought to you by Xebia Data.
• Travel and accommodation expenses are not included.