Research map / Label-efficient and robust learning / Recognition and long-tail
Towards Open Set Deep Networks
Check 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 |
|---|---|---|---|
| Add activation-vector Meta-Recognition and an unknown class to SoftMax | BVLC AlexNet · ImageNet 2012 val + open-set and fooling images | accuracy versus SoftMax with optimal threshold: 4.3% | Helped |
| Add activation-vector Meta-Recognition and an unknown class to SoftMax | BVLC AlexNet · ImageNet 2012 val + open-set and fooling images | accuracy versus base deep network: 12.3% | Helped |
| Use OpenMax rather than thresholding SoftMax probabilities | BVLC AlexNet · ImageNet 2012 val + open-set and fooling images | F-measure: 3450 more images | Helped |
| Use OpenMax rather than the base deep network | BVLC AlexNet · ImageNet 2012 val + open-set and fooling images | F-measure: 9847 more images | Helped |
| Use a weighted normalized Euclidean and cosine distance rather than pure Euclidean or other distance measures (ablation) | OpenMax · ImageNet 2012 | open-set recognition performance | No clear effect |
| Add uncertainty thresholding alongside the activation-vector unknown estimate (ablation) | OpenMax · ImageNet 2012 val + open-set and fooling images | open-set recognition | Helped |
Closest papers by meaning
- Learning Network Architecture for Open-Set Recognition (AAAI 2022 · similarity 0.90)
- GHOST: Gaussian Hypothesis Open-Set Technique (AAAI 2025 · similarity 0.86)
- Unlocking Better Closed-Set Alignment Based on Neural Collapse for Open-Set Recognition (AAAI 2025 · similarity 0.86)
- All Beings Are Equal in Open Set Recognition (AAAI 2024 · similarity 0.85)
- Frequency Shuffling and Enhancement for Open Set Recognition (AAAI 2024 · similarity 0.84)
- EGonc : Energy-based Open-Set Node Classification with substitute Unknowns (NeurIPS 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.
