Research map / Learning theory and algorithms / Prediction and decision losses
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 omniscient iterative teaching instead of random-example SGD | linear models on Gaussian data · Gaussian data | convergence speed | Helped |
| Use omniscient teacher instead of surrogate teacher | linear models in the same feature space · Gaussian data | convergence speed | Helped |
| Feed teacher-selected examples in sequence rather than running SGD on the selected set | linear models · Gaussian data | convergence speed | Helped |
| Use surrogate teaching in a different feature space | SVM · Gaussian data, different feature spaces | convergence | Hurt |
| Use imitation teacher instead of surrogate teacher | linear models in different feature spaces · Gaussian data | convergence | Helped |
| Apply teacher-guided example selection | linear classifiers · MNIST, 0/1 and 3/5 classification | convergence speed | Helped |
| Use imitation teacher instead of omniscient teacher | linear classifier · MNIST, 0/1 classification | convergence speed | No clear effect |
| Apply teacher-guided example selection to the CNN-6 fully connected layer | CNN-6 · CIFAR-10 | testing accuracy | Helped |
| Use imitation teacher instead of omniscient teacher | CNN-6 fully connected layer · CIFAR-10 | training objective value | Mixed |
| Use omniscient teacher instead of random-input SGD | learner on infant egocentric visual data · infant egocentric visual data | convergence speed | Helped |
Closest papers by meaning
- Iterative Teaching by Label Synthesis (NeurIPS 2021 · similarity 0.90)
- Towards Black-box Iterative Machine Teaching (ICML 2018 · similarity 0.89)
- Iterative Classroom Teaching (AAAI 2019 · similarity 0.89)
- Nonparametric Iterative Machine Teaching (ICML 2023 · similarity 0.88)
- Gradient-Based Algorithms for Machine Teaching (CVPR 2021 · similarity 0.88)
- Iterative Teacher-Aware Learning (NeurIPS 2021 · 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.