Accuracy (CS): 59%Accuracy (CV I): 26%Accuracy (CV II): 68%Accuracy (AV I): 43%
Papers with Code · CC BY-SA 4.0
Zero-Shot Learning
Zero-shot learning (ZSL) is a model's ability to detect classes never seen during training. The condition is that the classes are not known during supervised learning.
Earlier work in zero-shot learning use attributes in a two-step approach to infer unknown classes. In the computer vision context, more recent advances learn mappings from image feature space to semantic space. Other approaches learn non-linear multimodal embeddings. In the modern NLP context, language models can be evaluated on downstream tasks without fine tuning.
Benchmark datasets for zero-shot learning include aPY, AwA, and CUB, among others.
( Image credit: Prototypical Networks for Few shot Learning in PyTorch )
Further readings:
논문
Accuracy (CS): 60%Accuracy (CV I): 13%Accuracy (CV II): 33%Accuracy (AV I): 33%
Accuracy: 88.51%
Accuracy (CS): 76%Accuracy (CV I): 29%Accuracy (CV II): 71%Accuracy (AV I): 57%
Acc: 85.99Jaccard Index: 67.88F1-Score: 80.05
Accuracy: 86.1%
Average accuracy of 3 splits: 56.8
Accuracy: 94.77
Accuracy: 94%
Accuracy: 74%
Accuracy: 86.1Average accuracy of 3 splits: 86.1No. parameters: 1.75
Accuracy (Cross-Subject): 96.2
Accuracy: 91.32
Accuracy: 89.09±1.21
Accuracy (CS): 63%Accuracy (CV I): 14%Accuracy (CV II): 48%Accuracy (AV I): 48%
Accuracy: 78.0 (average of 3 split train/test)Average accuracy of 3 splits: 77.2No. parameters: 1.82 M
Average accuracy of 3 splits: 68.55
Average accuracy of 3 splits: 67.9No. parameters: -
Accuracy: 91.73
Average accuracy of 3 splits: 65.5
Actions Top-1 (S1): 39.2
Accuracy (CS): 71%Accuracy (CV I): 25%Accuracy (CV II): 56%Accuracy (AV I): 53%
Accuracy (CS): 56%Accuracy (CV I): 16%Accuracy (CV II): 43%Accuracy (AV I): 43%
Acc: 87.24Jaccard Index: 67.00F1-Score: 78.48
Accuracy: 52.3
Accuracy: 81.4%
Accuracy: 17.7%