Generalized Few-Shot Learning on ImageNet All classes 2012
80.2Top-5 AccuracyDynamic FSL + TRAML
Evaluation Results
| Method | Links | |
|---|---|---|
| Dynamic FSL + TRAMLNumber of shots (ns)=20, Hallucination=false, Backbone=ResNet102020.05 | 80.2 | |
| Dynamic FSLNumber of shots (ns)=20, Hallucination=false, Backbone=ResNet102020.05 | 78.7 | |
| Batch Squared Gradient MagnitudeNumber of shots (ns)=20, Hallucination=false, Backbone=ResNet102020.05 | 78.5 | |
| Squared Gradient Magnitude w/HNumber of shots (ns)=20, Hallucination=true, Backbone=ResNet102020.05 | 78.1 | |
| Prototype Matching Nets w/HNumber of shots (ns)=20, Hallucination=true, Backbone=ResNet102020.05 | 77.5 | |
| Dynamic FSL + TRAMLNumber of shots (ns)=10, Hallucination=false, Backbone=ResNet102020.05 | 77.3 | |
| Prototype Matching NetsNumber of shots (ns)=20, Hallucination=false, Backbone=ResNet102020.05 | 77.1 | |
| Logistic regressionNumber of shots (ns)=20, Hallucination=false, Backbone=ResNet102020.05 | 76.9 | |
| Logistic regression w/HNumber of shots (ns)=20, Hallucination=true, Backbone=ResNet102020.05 | 76.9 | |
| Matching NetworksNumber of shots (ns)=20, Hallucination=false, Backbone=ResNet102020.05 | 76.5 | |
| Dynamic FSLNumber of shots (ns)=10, Hallucination=false, Backbone=ResNet102020.05 | 76.5 | |
| Squared Gradient Magnitude w/HNumber of shots (ns)=10, Hallucination=true, Backbone=ResNet102020.05 | 75.8 | |
| Batch Squared Gradient MagnitudeNumber of shots (ns)=10, Hallucination=false, Backbone=ResNet102020.05 | 75.8 | |
| Prototype Matching Nets w/HNumber of shots (ns)=10, Hallucination=true, Backbone=ResNet102020.05 | 75.2 | |
| Prototype Matching NetsNumber of shots (ns)=10, Hallucination=false, Backbone=ResNet102020.05 | 75 | |
| Prototypical NetworkNumber of shots (ns)=20, Hallucination=false, Backbone=ResNet102020.05 | 74.6 | |
| Matching NetworksNumber of shots (ns)=10, Hallucination=false, Backbone=ResNet102020.05 | 73.7 | |
| Dynamic FSL + TRAMLNumber of shots (ns)=5, Hallucination=false, Backbone=ResNet102020.05 | 73.6 | |
| Prototypical NetworkNumber of shots (ns)=10, Hallucination=false, Backbone=ResNet102020.05 | 72.9 | |
| Logistic regression w/HNumber of shots (ns)=10, Hallucination=true, Backbone=ResNet102020.05 | 72.8 | |
| Dynamic FSLNumber of shots (ns)=5, Hallucination=false, Backbone=ResNet102020.05 | 72.2 | |
| Logistic regressionNumber of shots (ns)=10, Hallucination=false, Backbone=ResNet102020.05 | 71.9 | |
| Prototype Matching Nets w/HNumber of shots (ns)=5, Hallucination=true, Backbone=ResNet102020.05 | 71.9 | |
| Batch Squared Gradient MagnitudeNumber of shots (ns)=5, Hallucination=false, Backbone=ResNet102020.05 | 71.4 | |
| Squared Gradient Magnitude w/HNumber of shots (ns)=5, Hallucination=true, Backbone=ResNet102020.05 | 71.3 | |
| Prototype Matching NetsNumber of shots (ns)=5, Hallucination=false, Backbone=ResNet102020.05 | 71.1 | |
| Prototypical NetworkNumber of shots (ns)=5, Hallucination=false, Backbone=ResNet102020.05 | 69.7 | |
| Matching NetworksNumber of shots (ns)=5, Hallucination=false, Backbone=ResNet102020.05 | 69 | |
| Logistic regression w/HNumber of shots (ns)=5, Hallucination=true, Backbone=ResNet102020.05 | 67.6 | |
| Dynamic FSL + TRAMLNumber of shots (ns)=2, Hallucination=false, Backbone=ResNet102020.05 | 66.2 | |
| Dynamic FSLNumber of shots (ns)=2, Hallucination=false, Backbone=ResNet102020.05 | 65.2 | |
| Prototype Matching Nets w/HNumber of shots (ns)=2, Hallucination=true, Backbone=ResNet102020.05 | 64.7 | |
| Logistic regressionNumber of shots (ns)=5, Hallucination=false, Backbone=ResNet102020.05 | 64.2 | |
| Prototype Matching NetsNumber of shots (ns)=2, Hallucination=false, Backbone=ResNet102020.05 | 63.1 | |
| Squared Gradient Magnitude w/HNumber of shots (ns)=2, Hallucination=true, Backbone=ResNet102020.05 | 62.1 | |
| Prototypical NetworkNumber of shots (ns)=2, Hallucination=false, Backbone=ResNet102020.05 | 61 | |
| Matching NetworksNumber of shots (ns)=2, Hallucination=false, Backbone=ResNet102020.05 | 61 | |
| Batch Squared Gradient MagnitudeNumber of shots (ns)=2, Hallucination=false, Backbone=ResNet102020.05 | 60.5 | |
| Logistic regression w/HNumber of shots (ns)=2, Hallucination=true, Backbone=ResNet102020.05 | 59.4 | |
| Dynamic FSL + TRAMLNumber of shots (ns)=1, Hallucination=false, Backbone=ResNet102020.05 | 59.2 | |
| Dynamic FSLNumber of shots (ns)=1, Hallucination=false, Backbone=ResNet102020.05 | 58.2 | |
| Prototype Matching Nets w/HNumber of shots (ns)=1, Hallucination=true, Backbone=ResNet102020.05 | 57.6 | |
| Prototype Matching NetsNumber of shots (ns)=1, Hallucination=false, Backbone=ResNet102020.05 | 55.8 | |
| Matching NetworksNumber of shots (ns)=1, Hallucination=false, Backbone=ResNet102020.05 | 54.4 | |
| Squared Gradient Magnitude w/HNumber of shots (ns)=1, Hallucination=true, Backbone=ResNet102020.05 | 54.3 | |
| Logistic regression w/HNumber of shots (ns)=1, Hallucination=true, Backbone=ResNet102020.05 | 52.2 | |
| Logistic regressionNumber of shots (ns)=2, Hallucination=false, Backbone=ResNet102020.05 | 49.9 | |
| Prototypical NetworkNumber of shots (ns)=1, Hallucination=false, Backbone=ResNet102020.05 | 49.5 | |
| Batch Squared Gradient MagnitudeNumber of shots (ns)=1, Hallucination=false, Backbone=ResNet102020.05 | 49.3 | |
| Logistic regressionNumber of shots (ns)=1, Hallucination=false, Backbone=ResNet102020.05 | 40.8 |