References

Core methods

  • Lee & Schuler (2025). RieszBoost: Gradient Boosting for Riesz Regression. arXiv:2501.04871 — the algorithm rieszboost implements.
  • Singh (2021). Kernel Methods for Unobserved Confounding: Negative Controls, Proxies, and Instruments. arXiv:2102.11076 — closed-form RKHS estimator that krrr extends to the full set of estimands.

Riesz regression background

  • Williams, Hines & Rudolph (2025). Riesz representers for the rest of us. arXiv:2507.19413 — accessible introduction to the Riesz representation theorem and its role in semiparametric efficient estimation.
  • Chernozhukov, Newey, Quintas-Martínez & Syrgkanis (2021). RieszNet and ForestRiesz: Automatic Debiased Machine Learning of Causal and Structural Effects. arXiv:2110.03031 — neural-net and random-forest Riesz regression.
  • Chernozhukov, Newey, Quintas-Martínez & Syrgkanis (2022). Automatic Debiased Machine Learning via Riesz Regression. arXiv:2104.14737 — origin of the squared Riesz loss.

Bregman generalization

Beyond linear functionals

  • van der Laan, Luedtke & van der Laan (2025). Automatic Doubly-Robust Estimation for Smooth Functionals of the Outcome Mechanism. arXiv:2501.11868 — auto-DML for smooth functionals beyond linear.

Solver foundations (krrr)

  • Rudi, Carratino & Rosasco (2017). FALKON: An Optimal Large Scale Kernel Method. arXiv:1705.10958 — Nyström + preconditioned CG, optional GPU backend.
  • Rahimi & Recht (2007). Random Features for Large-Scale Kernel Machines. NIPS — basis of the random Fourier features solver.

Semiparametric estimation context

  • Hubbard, van der Laan & Robins (2011). Estimation of the Causal Effects of Time-Varying Exposures. The delta-method-EIF reference for partial-parameter estimands like ATT and LASE.

Software

The full set of papers (with arXiv source archives) lives at reference/ at the top level of the repo.