Research map / Learning theory and algorithms / Sample complexity
Distribution-Independent PAC Learning of Halfspaces with Massart Noise
Ilias Diakonikolas, Themis Gouleakis, Christos Tzamos
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 |
|---|---|---|---|
| Apply convex proxy optimization iteratively on high-margin regions, then classify the remaining points recursively | Halfspace learning with Massart noise · Distribution-independent Massart noise | misclassification error: \eta+\epsilon | Improved |
| Minimize a convex surrogate in one shot over the entire space | Halfspace learning with Massart noise · Large-margin halfspaces | misclassification error: \min\{\frac{\eta}{8\gamma},\frac12\} | Hurt |
| Use distribution preprocessing to remove outliers and rescale the retained data before applying the iterative learner | Iterative thresholding with LeakyReLU proxy · Arbitrary bit-complexity distributions with Massart noise | misclassification error: \eta+\epsilon | Improved |
Closest papers by meaning
- Learning Noisy Halfspaces with a Margin: Massart is No Harder than Random (NeurIPS 2024 · similarity 0.93)
- A Near-optimal Algorithm for Learning Margin Halfspaces with Massart Noise (NeurIPS 2024 · similarity 0.93)
- An Efficient Tester-Learner for Halfspaces (ICLR 2024 · similarity 0.91)
- Reliable Learning of Halfspaces under Gaussian Marginals (NeurIPS 2024 · similarity 0.91)
- Attribute-Efficient Learning of Sparse Halfspaces with Constant Malicious Noise Rate (NeurIPS 2026 · similarity 0.89)
- Smoothed Agnostic Learning of Halfspaces over the Hypercube (NeurIPS 2025 · similarity 0.87)
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.
