Research map / Learning theory and algorithms / Sample complexity
Improved Algorithms for Agnostic Pool-based Active Classification
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 ACED's experimental-design sampling instead of the evaluated baseline strategies | Binary image classification · MNIST 0-4 vs 5-9 | queries to reach passive accuracy | Helped |
| Use ACED's experimental-design sampling instead of the evaluated baseline strategies | Binary image classification · SVHN 2 vs 7 | queries to reach passive accuracy | Helped |
| Use ACED's experimental-design sampling instead of the evaluated baseline strategies | Binary image classification · CIFAR Bird vs Plane | queries to reach passive accuracy | Helped |
| Use ACED's experimental-design sampling instead of the evaluated baseline strategies | Binary image classification · FashionMNIST T-shirt vs Pants | queries to reach passive accuracy | Helped |
| Use OAC with a higher query-aggressiveness hyperparameter C0 | OAC · Image classification datasets | accuracy | Hurt |
| Use the online learner rather than retraining the classifier to convergence in batches | Active learning implementations · Image classification datasets | accuracy: <1% | Mixed |
| Use baseline IWAL algorithms on a linearly separable task | IWAL variants · FashionMNIST T-shirt vs Pants | accuracy | No clear effect |
Closest papers by meaning
- A Competitive Algorithm for Agnostic Active Learning (NeurIPS 2023 · similarity 0.90)
- Active Classification with Few Queries under Misspecification (NeurIPS 2024 · similarity 0.88)
- Querying Easily Flip-flopped Samples for Deep Active Learning (ICLR 2024 · similarity 0.86)
- Active Learning of General Halfspaces: Label Queries vs Membership Queries (NeurIPS 2024 · similarity 0.86)
- Universal Rates for Active Learning (NeurIPS 2024 · similarity 0.86)
- Neural Active Learning Beyond Bandits (ICLR 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.