Research map / Label-efficient and robust learning / Recognition and long-tail
One-Pass Learning with Incremental and Decremental Features
Chenping Hou, Zhi‐Hua Zhou
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add the C-stage classifier compressing vanished features (OPID vs SVM on all features) (ablation) | linear classifier · Mnist 0vs5 (n_i=40) | test accuracy: .9485 -> .9840 | Improved |
| add the C-stage classifier (OPID vs SVM) (ablation) | linear classifier · DNA (n_i=60) | test accuracy: .7693 -> .9253 | Improved |
| add the C-stage classifier (OPID vs SVM) (ablation) | linear classifier · Splice (n_i=40) | test accuracy: .6625 -> .8025 | Improved |
| add the C-stage classifier (OPID vs SVM) (ablation) | linear classifier · SensIT Vehicle (n_i=60) | test accuracy: .6533 -> .7147 | Improved |
| add the C-stage classifier (OPID vs SVM) (ablation) | linear classifier · Gisette (n_i=40) | test accuracy: .8795 -> .9710 | Improved |
| add C-stage information versus survived features only (OPID vs SVM(s)) (ablation) | linear classifier · DNA (n_i=60) | test accuracy: .8017 -> .9253 | Improved |
| expand with augmented features while inheriting C-stage metric (OPID vs SVM(a)) (ablation) | linear classifier · Mnist 0vs5 (n_i=40) | test accuracy: .7105 -> .9840 | Improved |
| ensemble implementation OPIDe instead of OPID (ablation) | OPID · Mnist 0vs5 (n_i=40) | test accuracy: .9840 -> .9700 | No clear effect |
| ensemble implementation OPIDe instead of OPID (ablation) | OPID · Gisette (n_i=300) | test accuracy: .9533 -> .9636 | Improved |
| ensemble implementation OPIDe instead of OPID (ablation) | OPID · DNA (n_i=300) | test accuracy: .9348 -> .9405 | Improved |
| increase E-stage training points from 40 to 320 | SVM · Mnist 0vs5 | test accuracy: .9485 -> .9604 | Improved |
| increase survived-feature percentage (ablation) | OPID and OPIDe · SensIT Vehicle | test accuracy | Improved |
Closest papers by meaning
- Learning with Feature Evolvable Streams (NeurIPS 2017 · similarity 0.84)
- Prototype Augmentation and Self-Supervision for Incremental Learning (CVPR 2021 · similarity 0.82)
- eTag: Class-Incremental Learning via Embedding Distillation and Task-Oriented Generation (AAAI 2024 · similarity 0.82)
- One-Pass Feature Evolvable Learning with Theoretical Guarantees (ICML 2025 · similarity 0.82)
- Adaptive Decision Boundary for Few-Shot Class-Incremental Learning (AAAI 2025 · similarity 0.81)
- Adaptive Retention & Correction: Test-Time Training for Continual Learning (ICLR 2025 · similarity 0.80)
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.
