Research map / Label-efficient and robust learning / Recognition and long-tail
Joint Learning of Set Cardinality and State Distribution
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 |
|---|---|---|---|
| Jointly learn cardinality and label distributions with shared weights instead of training them independently | Deep set network with BCE-DC · PASCAL VOC 2007 | C-F1: 76.2 -> 78.7 | Helped |
| Jointly learn cardinality and label distributions with shared weights instead of training them independently | Deep set network with BCE-DC · PASCAL VOC 2007 | O-F1: 79.0 -> 81.5 | Helped |
| Jointly learn cardinality and label distributions with shared weights instead of training them independently | Deep set network with BCE-DC · PASCAL VOC 2007 | I-F1: 83.0 -> 85.1 | Helped |
| Replace negative binomial cardinality loss with Dirichlet-Categorical (ablation) | Independent deep set network · PASCAL VOC 2007 | C-F1: 75.8 -> 76.2 | Helped |
| Replace negative binomial cardinality loss with Dirichlet-Categorical (ablation) | Independent deep set network · MS COCO | C-F1: 64.6 -> 64.7 | Helped |
| Replace negative binomial cardinality loss with Dirichlet-Categorical (ablation) | Independent deep set network · MS COCO | O-F1: 69.4 -> 69.6 | Helped |
| Replace negative binomial cardinality loss with Dirichlet-Categorical (ablation) | Independent deep set network · MS COCO | I-F1: 74.4 -> 74.6 | Helped |
| Replace negative binomial cardinality loss with Dirichlet-Categorical (ablation) | Independent deep set network · PASCAL VOC 2007 | O-F1: 78.6 -> 79.0 | Helped |
| Replace negative binomial cardinality loss with Dirichlet-Categorical (ablation) | Independent deep set network · PASCAL VOC 2007 | I-F1: 82.6 -> 83.0 | Helped |
| Learn cardinality jointly rather than independently | Cardinality estimator · PASCAL VOC 2007 | mean absolute cardinality error: 0.33±0.53 -> 0.31±0.54 | Helped |
| Train classification with the cardinality term jointly rather than using BCE alone (ablation) | Classifier · PASCAL VOC 2007 | classification precision/recall | Helped |
Closest papers by meaning
- Deep Set Prediction Networks (NeurIPS 2019 · similarity 0.87)
- Set Prediction in the Latent Space (NeurIPS 2021 · similarity 0.87)
- Multiset-Equivariant Set Prediction with Approximate Implicit Differentiation (ICLR 2022 · similarity 0.86)
- Predict and Constrain: Modeling Cardinality in Deep Structured Prediction (ICML 2018 · similarity 0.86)
- Powerset Convolutional Neural Networks (NeurIPS 2019 · similarity 0.86)
- Cardinality-Aware Set Prediction and Top-$k$ Classification (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.