Research map / Label-efficient and robust learning / Recognition and long-tail
Towards Black-box Iterative Machine 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 |
|---|---|---|---|
| Use active queries to estimate the learner and choose teaching examples instead of feeding random examples | Cross-space iterative machine teaching · Synthetic linear regression and binary classification | convergence | Helped |
| Recover the virtual learner exactly using one-time examination feedback | Cross-space iterative machine teaching · Synthetic regression and logistic regression with sigmoid feedback | convergence | Helped |
| Use approximate recovery from sign-only feedback rather than exact recovery | Cross-space iterative machine teaching · Synthetic logistic regression with sign feedback | convergence | Mixed |
| Use approximate recovery from sign-only feedback rather than exact recovery | Cross-space iterative machine teaching · MNIST 7/9 classification | accuracy | Mixed |
| Use approximate recovery from sign-only feedback rather than exact recovery | Cross-space iterative machine teaching · MNIST 7/9 classification | convergence | Hurt |
| Restrict teaching to a fixed pool with rescaling while using non-bijective feedback | Active teacher · Rescalable pool-based teaching | exponential teachability | Hurt |
| Use one initial examination and propagate the virtual learner without further exams | Active teacher with unitary feature mapping · Cross-space teaching | query complexity | Helped |
| Provide the norm of the learner's parameters in addition to sign feedback | Active teacher using one-bit feedback · Cross-space teaching | virtual learner recovery | Helped |
Closest papers by meaning
- Iterative Machine Teaching (ICML 2017 · similarity 0.89)
- Teaching with Limited Information on the Learner’s Behaviour (ICML 2020 · similarity 0.88)
- Iterative Classroom Teaching (AAAI 2019 · similarity 0.87)
- Nonparametric Iterative Machine Teaching (ICML 2023 · similarity 0.86)
- Iterative Teacher-Aware Learning (NeurIPS 2021 · similarity 0.86)
- Machine Teaching of Active Sequential Learners (NeurIPS 2019 · similarity 0.86)
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.