An Introduction to 'metalearners', a Flexible MetaLearner Library in Python
summary
A Python library called metalearners implements R‑, X‑ and DR‑style meta‑learners that output CATE estimates via a fit/predict API, reuses fitted base models and runs hyperparameter search with Optuna; examples use LightGBM as base learners and SHAP for explanations. A simulated experiment shows heterogeneity in individual treatment effects (histogram crosses zero) and the code fits in a few lines, reports per‑component predictions, and supports pandas/numpy/ONNX workflows from QuantCo on GitHub.
AI-written summary · about
chapters (7)
- 0:00Introduction to metalearners
- 3:17Pill choice problem setup
- 8:48Fundamental causal inference problem
- 12:00MetaLearners architecture
- 17:22Hyperparameter optimization
- 22:41Real-world applications
- 24:35Q&A
related talks
50mFlexible Neural Networks and the Frontiers of Meta-Learning
10mMIND MELD: Personalized Meta-Learning for Robot-Centric Imitation Learning
57mGood Practices for Applied Machine Learning- 57mGood Practices for Applied Machine Learning
44mMachine Learning Made Easy With PyCaret
46mIncreasing Participant Agency by Understanding the Impact of Training Data
48mFKLearn, A functional machine learning library
15mInterpretable Machine Learning with R