Lune

Abstract

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--along with any consequent causal changes in other predictor variables\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.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Related papers

  1. Explaining Hyperparameter Optimization via Partial Dependence PlotsJulia Moosbauer, Julia Herbinger, Giuseppe Casalicchio, Marius Lindauer et al.NeurIPS 2021 · 109 citations
  2. Mind the Graph When Balancing Data for Fairness or RobustnessJessica Schrouff, Alexis Bellot, Amal Rannen-Triki, Alan Malek et al.NeurIPS 2024 · 10 citations
  3. Enhancing LLM Agent Safety via Causal Influence PromptingDongyoon Hahm, Woogyeol Jin, June Suk Choi, Sungsoo Ahn et al.ACL Findings 2025 · 1 citation
  4. DeciX: Explain Deep Learning Based Code Generation ApplicationsSimin Chen, Zexin Li, Wei Yang, Cong LiuFSE 2024 · 1 citation
  5. Sense and Sensitivity Analysis: Simple Post-Hoc Analysis of Bias Due to Unobserved ConfoundingVictor Veitch, Anisha ZaveriNeurIPS 2020 · 67 citations
Dusk over the sea between two cliffs drawn in fine vertical lines