Causal Dependence Plots
This provides a tool for interpretable machine learning, aiding in applications like scientific ML and fairness, but it is incremental as it builds on existing causal visualization concepts.
The authors tackled the problem of explaining AI/ML models by developing Causal Dependence Plots (CDPs) to visualize how an outcome depends on a predictor while accounting for causal changes in other variables, showing their modular integration with causal learning methods.
Explaining artificial intelligence or machine learning models is increasingly important. To use such data-driven systems wisely we must understand how they interact with the world, including how they depend causally on data inputs. In this work we develop Causal Dependence Plots (CDPs) to visualize how one variable--an outcome--depends on changes in another variable--a predictor--$\textit{along with any consequent causal changes in other predictor variables}$. Crucially, CDPs differ from standard methods based on holding other predictors constant or assuming they are independent. CDPs make use of an auxiliary causal model because causal conclusions require causal assumptions. With simulations and real data experiments, we show CDPs can be combined in a modular way with methods for causal learning or sensitivity analysis. Since people often think causally about input-output dependence, CDPs can be powerful tools in the xAI or interpretable machine learning toolkit and contribute to applications like scientific machine learning and algorithmic fairness.