SHAP (SHapley Additive exPlanations) values for Interpretable Machine Learning
Day 2 Session 2 (1:30 PM)
Presenters:
Ali Dadkhah, University of Vermont
Harrison Myers, University of Vermont
Shaurya Swami, University of Vermont
Ryan van der Heijden, University of Vermont
Kristen Underwood, PhD, University of Vermont
This workshop introduces participants to SHAP (SHapley Additive exPlanations) values (Lundberg & Lee, 2017), a useful tool for interpreting machine learning models by analyzing feature importance. Participants will learn the fundamentals of SHAP values and their practical applications to understand how individual features influence predictions in regression and classification models. Using Python libraries, we will walk through hands-on examples to explore how SHAP values can enhance model transparency, provide actionable insights, and build trust in machine learning predictions, with a focus on real-world applications in water resources.
Learning Outcomes:
- Understand SHAP values and their role in interpreting machine learning models.
- How to implement SHAP values in Python
- How to post-process SHAP tool output to create visualizations (e.g., bee swarm plot)
- How qualitative evaluation of SHAP values can yield process insights for the system being modeled
Prerequisites:
Knowledge:
- Python coding knowledge
- Fundamentals of statistics including multivariate statistics and factor interactions.
Hardware/Software:
- GitHub – shap/shap: A game theoretic approach to explain the output of any machine learning model.
Accounts:
- HydroShare: Participants will need to request access to join the DevCon group or CUAHSI JupyterHub group