Interpretability in H2O Driverless AI

Overview

H2O Driverless AI provides robust interpretability of machine learning models to explain modeling results with the Machine Learning Interpretability (MLI) capability. MLI enables a data scientist to employ different techniques and methodologies for interpreting and explaining the results of its models with four charts that are generated automatically including: K-LIME, Shapley, Variable Importance, Decision Tree, Partial Dependence and more.

Key Capabilities of Our Machine Learning Interpretability

  • Shapley
  • k-LIME
  • Surrogate Decision Trees
  • Partial Dependence Plot
  • LOCO

Local Shapley feature importance shows how each feature directly impacts each individual row’s prediction. Local Shapley values are accurate, consistent, and likely suitable for creating reason codes and adverse action codes and always add up to the model prediction.

Global Shapley feature importance provides an overall view of the drivers of your model’s predictions. Global Shapley values are reported for original features and any feature the Driverless AI system creates on its own.

k-LIME automatically builds linear model approximations to regions of complex Driverless AI models’ learned response functions. These penalized GLM surrogates are trained to model the predictions of the Driverless AI model and can be used to generate reason codes and English language explanations of complex Driverless AI models

The local surrogate decision tree path shows how the logic of the model is applied to any given individual. The decision paths hows approximately how row values impact Driverless AI model predictions for that row by showing the row’s path through the decision tree surrogate flow chart. The global surrogate decision tree provides an overall flowchart of the Driverless AI model’s  decision making  processes based on the original features. The surrogate decision tree shows higher and more frequent features in the tree that are more important to the Driverless AI model than lower or less frequent variables.

Partial dependence shows the average Driverless AI model prediction and its standard deviation for different values of important original features. This helps you understand the average model behavior for the most important original features.

Local feature importance describes how the combination of the learned model rules or parameters and an individual row’s attributes affect a model’s prediction for that row while taking nonlinearity and interactions into effect. Local feature importance values reported here are based on a variant of the leave-one-covariate-out (LOCO) method (Lei et al, 2017).