Research map / Learning theory and algorithms / Sample complexity
Improved Algorithms for Collaborative PAC Learning
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 R1 instead of the previous centralized algorithm | Collaborative PAC learning · Realizable, centralized | sample complexity | Helped |
| Use R2's FastTest with more rounds instead of R1's higher-sample testing | Collaborative PAC learning · Realizable | sample complexity | Helped |
| Use Algorithm 2 rather than Algorithm 1 | Collaborative PAC learning · Realizable, typical values of δ | sample complexity | Helped |
| Use R2 when d=Θ(k) | Collaborative PAC learning · Realizable | sample complexity | Helped |
| Use NR1-AVG or NR2-AVG randomized classifiers instead of the non-randomized NR1 or NR2 classifiers | Collaborative PAC learning · Non-realizable | error approximation factor: 2+α -> 1+α | Helped |
| Use randomized classifiers to avoid the factor-2 error guarantee | Collaborative PAC learning · Non-realizable | sample complexity: O\Big(\frac{1}{\epsilon}\Big) times more samples | Hurt |
| Use Algorithm 2 rather than Algorithm 1 | Collaborative PAC learning · Non-realizable, (2+α) approximation | sample complexity | Helped |
| Use Algorithm 2 rather than Algorithm 1 | Collaborative PAC learning · Non-realizable, randomized expected-error guarantee | sample complexity | Mixed |
Closest papers by meaning
- Collaborative Learning with Different Labeling Functions (ICML 2024 · similarity 0.93)
- Fast Rates for Bandit PAC Multiclass Classification (NeurIPS 2024 · similarity 0.87)
- Improved Sample Complexity for Multiclass PAC Learning (NeurIPS 2024 · similarity 0.86)
- Fast Rate Bounds for Multi-Task and Meta-Learning with Different Sample Sizes (NeurIPS 2025 · similarity 0.83)
- Learning Distributions from Multiple Data Providers (NeurIPS 2026 · similarity 0.83)
- On the Computational Landscape of Replicable Learning (NeurIPS 2024 · 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.