MAP: 41.7
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
MAP: 89.2
H-Mean: 16.1Seen accuracy: 24.4Test AUC top 1: 3.0Test AUC top 2: 7.6
MAP: 73.6
Attribute accuracy: 30.2Object accuracy: 32.6Seen accuracy: 29Unseen accuracy: 26.2
mAP: 39.01
MAP: 62.9
H-Mean: 39.9Seen accuracy: 51.0Test AUC top 1: 23.4Test AUC top 2: -
Harmonic mean: 41.3
MAP: 41.2
Harmonic mean: 39.4
mAP: 25.7
MAP: 40.5
Harmonic mean: 40.2
mAP: 31.1
Harmonic mean: 39.4
Harmonic mean: 34.0
MAP: 40.4
F1: 45.21
H: 43.2