TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Papers/Generalized Adaptation for Few-Shot Learning

Generalized Adaptation for Few-Shot Learning

Liang Song, Jinlu Liu, Yongqiang Qin

2019-11-25Few-Shot LearningFew-Shot Image Classification
PaperPDF

Abstract

Many Few-Shot Learning research works have two stages: pre-training base model and adapting to novel model. In this paper, we propose to use closed-form base learner, which constrains the adapting stage with pre-trained base model to get better generalized novel model. Following theoretical analysis proves its rationality as well as indication of how to train a well-generalized base model. We then conduct experiments on four benchmarks and achieve state-of-the-art performance in all cases. Notably, we achieve the accuracy of 87.75% on 5-shot miniImageNet which approximately outperforms existing methods by 10%.

Results

TaskDatasetMetricValueModel
Image ClassificationCIFAR-FS 5-way (1-shot)Accuracy73.1ACC + Amphibian
Image ClassificationMini-Imagenet 5-way (5-shot)Accuracy80.75ACC + Amphibian
Image ClassificationMini-Imagenet 5-way (1-shot)Accuracy62.21ACC + Amphibian
Image ClassificationFC100 5-way (5-shot)Accuracy66.9ACC + Amphibian
Image ClassificationFC100 5-way (1-shot)Accuracy41.6ACC + Amphibian
Image ClassificationTiered ImageNet 5-way (1-shot)Accuracy68.77ACC + Amphibian
Image ClassificationTiered ImageNet 5-way (5-shot)Accuracy86.75ACC + Amphibian
Image ClassificationCIFAR-FS 5-way (5-shot)Accuracy89.3ACC + Amphibian
Few-Shot Image ClassificationCIFAR-FS 5-way (1-shot)Accuracy73.1ACC + Amphibian
Few-Shot Image ClassificationMini-Imagenet 5-way (5-shot)Accuracy80.75ACC + Amphibian
Few-Shot Image ClassificationMini-Imagenet 5-way (1-shot)Accuracy62.21ACC + Amphibian
Few-Shot Image ClassificationFC100 5-way (5-shot)Accuracy66.9ACC + Amphibian
Few-Shot Image ClassificationFC100 5-way (1-shot)Accuracy41.6ACC + Amphibian
Few-Shot Image ClassificationTiered ImageNet 5-way (1-shot)Accuracy68.77ACC + Amphibian
Few-Shot Image ClassificationTiered ImageNet 5-way (5-shot)Accuracy86.75ACC + Amphibian
Few-Shot Image ClassificationCIFAR-FS 5-way (5-shot)Accuracy89.3ACC + Amphibian

Related Papers

GLAD: Generalizable Tuning for Vision-Language Models2025-07-17ViT-ProtoNet for Few-Shot Image Classification: A Multi-Benchmark Evaluation2025-07-12Doodle Your Keypoints: Sketch-Based Few-Shot Keypoint Detection2025-07-10An Enhanced Privacy-preserving Federated Few-shot Learning Framework for Respiratory Disease Diagnosis2025-07-10Few-Shot Learning by Explicit Physics Integration: An Application to Groundwater Heat Transport2025-07-08ViRefSAM: Visual Reference-Guided Segment Anything Model for Remote Sensing Segmentation2025-07-03Dynamic Context-Aware Prompt Recommendation for Domain-Specific AI Applications2025-06-25Ancient Script Image Recognition and Processing: A Review2025-06-24