Causal Machine Learning: Prediction to Intervention

Production causal ML series covering structural causal models, backdoor adjustment, and Inverse Propensity Scoring for offline treatment effect estimation.


Standard supervised models fail when policy changes the data-generating process. This series covers structural causal models, the backdoor criterion, feature-flattening traps, and Inverse Propensity Scoring (IPS) to debias observational logs without live A/B tests—for ML engineers shipping counterfactual evaluation.




In This Series

  1. Why Machine Learning Demands Causal Inference
    [Series] Causal Machine Learning - 1. Rethinking the Predictive Paradigm
  2. Fixing Selection Bias in Enterprise Data with Inverse Propensity Scoring (IPS)
    [Series] Causal Machine Learning - 2. De-Biasing Historical Data Logs (without Expensive A/B Tests)
Causal Machine Learning: Prediction to Intervention

PART 01 / 02

Why Machine Learning Demands Causal Inference

[Series] Causal Machine Learning - 1. Rethinking the Predictive Paradigm

Machine LearningDeep LearningData SciencePython

Standard machine learning models excel at empirical risk minimization on static observational data, but fail catastrophically when an operational policy actively alters the data-generating distribution.

This advanced guide exposes the feature flattening trap—where models mistake downstream symptoms for upstream causes—and details how to deploy structurally invariant causal models using Judea Pearls Backdoor Criterion and Off-Policy Evaluation (OPE).

Why Machine Learning Demands Causal Inference

Kernel Labs | Kuriko IWAI | kuriko-iwai.com

Read more

PART 02 / 02

Fixing Selection Bias in Enterprise Data with Inverse Propensity Scoring (IPS)

[Series] Causal Machine Learning - 2. De-Biasing Historical Data Logs (without Expensive A/B Tests)

Machine LearningDeep LearningData SciencePython

Standard machine learning models fail when observational data is poisoned by selection bias—mistaking baseline confounding for true treatment impact.

Inverse Propensity Scoring (IPS) provides an offline mathematical simulator that transforms biased operational logs into synthetic Randomized Controlled Trials (RCTs).

This guide breaks down the underlying propensity mathematics, positivity constraints, and production Python code to calculate unconfounded Average Treatment Effects (ATE).

Fixing Selection Bias in Enterprise Data with Inverse Propensity Scoring (IPS)

Kernel Labs | Kuriko IWAI | kuriko-iwai.com

Read more

Shipping AI Systems?

I help teams design and deploy scalable ML / RAG / LLM pipelines and MLOps infrastructure.



Or explore: