# AIRL

**Reference PDF:** pending known-truth migration.

AIRL learns a reward through an adversarial discriminator with a potential-based
shaping term. It is part of the 12-estimator known-truth validation suite.

This RTD page is intentionally short while the dedicated AIRL TeX/PDF tutorial
is migrated to the shared synthetic DGP. No real-data estimation examples are
published on RTD.

## Migration Target

- PDF source: `papers/econirl_package/primers/airl/airl.tex`
- Result generator: `papers/econirl_package/primers/airl/airl_run.py`
- Shared DGP harness: `experiments/known_truth.py`