Research map / Architectures and efficiency / Pruning and efficient tuning
Automatic Channel Pruning by Searching with Structure Embedding for Hash Network
Zifan Liu, Yuan Cao, Yifan Sun, Yanwei Yu, Heng Qi
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 |
|---|---|---|---|
| use unsupervised contrastive learning instead of supervised mAP regression for GCN encoder (ablation) | ACP-SSE · CIFAR-10 DCH | mAP@all: 0.7623 -> 0.8215 | Improved |
| use unsupervised contrastive learning instead of supervised mAP regression for GCN encoder (ablation) | ACP-SSE · CIFAR-10 DTSH | mAP@all: 0.7832 -> 0.8387 | Improved |
| use unsupervised contrastive learning instead of supervised mAP regression for GCN encoder (ablation) | ACP-SSE · CIFAR-10 CSQ | mAP@all: 0.8041 -> 0.8521 | Improved |
| add top-k refinement (fine-tune several candidates) (ablation) | ACP-SSE · CIFAR-10 DCH | mAP@all: 0.7748 -> 0.8215 | Improved |
| add top-k refinement (fine-tune several candidates) (ablation) | ACP-SSE · CIFAR-10 DTSH | mAP@all: 0.7887 -> 0.8387 | Improved |
| add top-k refinement (fine-tune several candidates) (ablation) | ACP-SSE · CIFAR-10 CSQ | mAP@all: 0.8175 -> 0.8521 | Improved |
| increase k from 1 to 10 (m=50) (ablation) | ACP-SSE CSQ ResNet-50 · CIFAR-10 | mAP: 0.824 -> 0.852 | Improved |
| increase candidate pool m beyond 50 (ablation) | ACP-SSE CSQ ResNet-50 · CIFAR-10 | mAP: 0.848 -> 0.852 | No clear effect |
| prune with ACP-SSE vs ABCPruner | DTSH ResNet-50 · CIFAR-10 | mAP@all: 0.8394 -> 0.8387 | Hurt |
| prune with ACP-SSE vs AutoSculpt | DCH ResNet-50 · CIFAR-10 | mAP@all: 0.8164 -> 0.8215 | Improved |
| prune with ACP-SSE vs ABCPruner | DCH ResNet-50 · ImageNet100 | mAP@all: 0.6433 -> 0.6608 | Improved |
Closest papers by meaning
- Deep Hashing Network for Efficient Similarity Retrieval (AAAI 2016 · similarity 0.87)
- Pruning from Scratch (AAAI 2020 · similarity 0.87)
- Dynamic Structure Pruning for Compressing CNNs (AAAI 2023 · similarity 0.87)
- Network Pruning via Transformable Architecture Search (NeurIPS 2019 · similarity 0.87)
- Deep Supervised Hashing for Fast Image Retrieval (CVPR 2016 · similarity 0.87)
- Deep Quantization Network for Efficient Image Retrieval (AAAI 2016 · similarity 0.86)
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.
