Research map / Learning theory and algorithms / Sample complexity
Efficient Truncated Linear Regression with Unknown Noise Variance
Constantinos Costis Daskalakis, Patroklos Stefanou, Rui Yao, Emmanouil Zampetakis
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 the proposed estimator with unknown variance instead of OLS under left truncation | truncated linear regression · synthetic, 10-dim, sigma^2 in [1,10] | w and sigma^2 estimation error | Improved |
| proposed method vs prior method given known variance | truncated linear regression · synthetic, 10-dim, sigma^2 in [1,10] | w estimation error | No clear effect |
| increase number of retained samples k from 10 to 5000 | proposed estimator · synthetic right truncation | estimation error bound | Improved |
| proposed method vs prior known-variance method | truncated linear regression · semi-synthetic PM10, truncation C in [1.0,4.0] | L2 parameter distance and L1 variance distance | No clear effect |
| increase truncation parameter C to high values | both methods · semi-synthetic PM10 | estimation error | Mixed |
Closest papers by meaning
- Linear Regression with Unknown Truncation Beyond Gaussian Features (ICML 2026 · similarity 0.87)
- Efficient algorithms for linear regression with heteroskedastic errors (NeurIPS 2026 · similarity 0.87)
- Optimal Linear Regression Without a Variance (NeurIPS 2026 · similarity 0.86)
- Non-Asymptotic Guarantees for Robust Statistical Learning under Infinite Variance Assumption (JMLR 2023 · similarity 0.85)
- Information-Computation Tradeoffs for Noiseless Linear Regression with Oblivious Contamination (NeurIPS 2025 · similarity 0.85)
- Robust Sparse Regression with Non-Isotropic Designs (NeurIPS 2024 · similarity 0.85)
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.
