Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Revisiting Prototypical Network for Cross Domain Few-Shot Learning
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 |
|---|---|---|---|
| Add self-image local-global distillation to ProtoNet++ (ablation) | ProtoNet++ with ResNet-10 · CUB, 5-way 1-shot | accuracy: 40.34 -> 43.99 | Helped |
| Add cross-image local-global distillation (ablation) | LDP-net with self-image distillation · CUB, 5-way 1-shot | accuracy: 43.99 -> 44.04 | Helped |
| Add cross-image local-global distillation (ablation) | LDP-net with self-image distillation · EuroSAT, 5-way 5-shot | accuracy: 80.22 -> 80.07 | Hurt |
| Add cross-episode EMA distillation (ablation) | LDP-net with self-image and cross-image distillation · CUB, 5-way 1-shot | accuracy: 44.04 -> 47.70 | Helped |
| Add cross-episode EMA distillation (ablation) | LDP-net with self-image and cross-image distillation · CUB, 5-way 5-shot | accuracy: 64.01 -> 68.94 | Helped |
| Use local-global distillation with cross-image consistency and cross-episode EMA (ablation) | ProtoNet++ · CUB, 5-way 1-shot | accuracy: 40.34 -> 47.70 | Helped |
| Use local-global distillation with cross-image consistency and cross-episode EMA (ablation) | ProtoNet++ · EuroSAT, 5-way 1-shot | accuracy: 59.11 -> 63.70 | Helped |
| Use local-global distillation with cross-image consistency and cross-episode EMA (ablation) | ProtoNet++ · EuroSAT, 5-way 5-shot | accuracy: 81.44 -> 80.26 | Hurt |
| Train with LDP-net instead of ProtoNet++ | ResNet-10 ProtoNet++ · CUB, 5-way 1-shot | accuracy: 40.34 -> 47.70 | Helped |
| Train with LDP-net instead of ProtoNet++ | ResNet-10 ProtoNet++ · EuroSAT, 5-way 1-shot | accuracy: 59.11 -> 63.70 | Helped |
Closest papers by meaning
- Prototypical Networks for Few-shot Learning (NeurIPS 2017 · similarity 0.90)
- Cross-Level Distillation and Feature Denoising for Cross-Domain Few-Shot Classification (ICLR 2023 · similarity 0.89)
- Dynamic Distillation Network for Cross-Domain Few-Shot Recognition with Unlabeled Data (NeurIPS 2021 · similarity 0.89)
- Mind the Gap Between Prototypes and Images in Cross-domain Finetuning (NeurIPS 2024 · similarity 0.87)
- Pushing the Limit of Fine-Tuning for Few-Shot Learning: Where Feature Reusing Meets Cross-Scale Attention (AAAI 2024 · similarity 0.87)
- Simple Semantic-Aided Few-Shot Learning (CVPR 2024 · 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.