Research map / Probabilistic and scientific modeling / Treatment effects
Orthogonal Machine Learning: Power and Limitations
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use second-order orthogonal moments instead of first-order orthogonal moments | PLR treatment-effect estimation · Synthetic demand estimation (n=5000, p=1000, s=100) | bias: 2.78 -> 3. | Helped |
| Use second-order orthogonality rather than first-order orthogonality | PLR treatment-effect estimation · High-dimensional linear nuisance functions | maximum sparsity scaling: o({n^{1/2}}{/\log p}) -> o({n^{2/3}}{ /\log p}) | Helped |
| Use k-th order orthogonal moments instead of first-order orthogonality | Two-stage Z-estimation · General nuisance-function estimation | sufficient nuisance estimation rate for root-n inference: n^{-1/4} -> n^{-1/(2k+2)} | Helped |
| Use second-order orthogonal moments when treatment residuals are non-Gaussian | PLR treatment-effect estimation · PLR with non-Gaussian treatment residual | sufficient nuisance estimation rate for root-n inference: o(n^{-1/4}) -> o(n^{-1/6}) | Helped |
| Use second-order orthogonal moments when treatment residuals are conditionally Gaussian | PLR treatment-effect estimation · PLR with conditionally Gaussian treatment residual | non-degenerate root-n inference | Hurt |
Closest papers by meaning
- Stochastic Gradients under Nuisances (NeurIPS 2025 · similarity 0.84)
- Inverse M-Kernels for Linear Universal Approximators of Non-Negative Functions (NeurIPS 2024 · similarity 0.84)
- Generalization in Kernel Regression Under Realistic Assumptions (ICML 2024 · similarity 0.83)
- Kernel Regression in Structured Non-IID Settings: Theory and Implications for Denoising Score Learning (NeurIPS 2025 · similarity 0.83)
- Doubly Robust Proximal Causal Learning for Continuous Treatments (ICLR 2024 · similarity 0.83)
- Kernel conditional tests from learning-theoretic bounds (NeurIPS 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.