Research map / Learning theory and algorithms / Sample complexity
Preference-Based Batch and Sequential Teaching: Towards a Unified View of Models
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 |
|---|---|---|---|
| Allow preference functions to depend on the candidate, version space, and current hypothesis | Sequential teaching · Finite hypothesis classes | teaching complexity: O(VCD(ℋ, 𝒳)^2) -> O(VCD(ℋ, 𝒳)) | Helped |
| Use version-space-dependent preferences instead of a global ordering | Batch teaching · Warmuth hypothesis class | teaching dimension: 3 -> 2 | Helped |
| Use version-space-dependent preferences instead of a global ordering | Batch teaching · Example hypothesis class (Appendix) | teaching dimension: 2 -> 1 | Helped |
| Use sequential local preferences rather than version-space-dependent batch preferences | Teaching models · Some hypothesis classes | teaching complexity | Mixed |
| Use local preferences instead of version-space-dependent preferences | Teaching models · Power-set class with k=7 | teaching dimension: ≥ 4 -> ≤ 3 | Helped |
Closest papers by meaning
- On Batch Teaching with Sample Complexity Bounded by VCD (NeurIPS 2022 · similarity 0.90)
- Teaching with Limited Information on the Learner’s Behaviour (ICML 2020 · similarity 0.89)
- Machine Teaching of Active Sequential Learners (NeurIPS 2019 · similarity 0.88)
- Understanding the Role of Adaptivity in Machine Teaching: The Case of Version Space Learners (NeurIPS 2018 · similarity 0.87)
- On the Complexity of Teaching a Family of Linear Behavior Cloning Learners (NeurIPS 2024 · similarity 0.85)
- Unpacking DPO and PPO: Disentangling Best Practices for Learning from Preference Feedback (NeurIPS 2024 · similarity 0.84)
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.