Research map / Label-efficient and robust learning / Knowledge distillation
Iterative Classroom Teaching
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 |
|---|---|---|---|
| Choose shared examples using the leading eigenvector of the learners' weighted error matrix | Classroom teaching for online projected gradient descent learners · Iterative classroom teaching | teaching examples: \mathcal{O} \br{N \log \frac{1}{\epsilon}} -> \mathcal{O} \br{k \log \frac{1}{\epsilon}} | Helped |
| Use noisy estimates of learner states to construct examples | Classroom teaching · Noisy student-state observations | teaching rounds: \mathcal{O}\br{\br{\log \frac{1}{1 - \frac{\alpha_{\mathrm{min}}}{k}}}^{-1} \log \frac{1}{\epsilon}} | Helped |
| Estimate time-varying learning rates from past values | Classroom teaching · Time-varying learning rates | expected teaching rounds: \mathcal{O}\br{\br{\log \frac{1}{1 - \frac{\beta_{\mathrm{min}} \bar \alpha_{\mathrm{min}}}{k}}}^{-1} \log \frac{1}{\eps | Helped |
| Partition learners into homogeneous groups based on learning rate | CT · Synthetic learners with diverse learning rates | teacher examples to convergence | Mixed |
| Partition learners by prior knowledge or hypothesis type | CT · Synthetic learners with diverse initial states; butterfly and moth classification | student workload | Helped |
| Use dynamic rather than static example magnitude | Classroom teaching · Iterative classroom teaching | sample complexity | Mixed |
Closest papers by meaning
- Iterative Machine Teaching (ICML 2017 · similarity 0.89)
- Teaching with Limited Information on the Learner’s Behaviour (ICML 2020 · similarity 0.89)
- On a Combinatorial Problem Arising in Machine Teaching (ICML 2024 · similarity 0.85)
- Cooperative Knowledge Distillation: A Learner Agnostic Approach (AAAI 2024 · similarity 0.85)
- On the Complexity of Teaching a Family of Linear Behavior Cloning Learners (NeurIPS 2024 · similarity 0.85)
- Toward Student-oriented Teacher Network Training for Knowledge Distillation (ICLR 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.