The Three Stage
This paper combines SimSiam(a self-supervised learning method to learn feature representations) with active learning to reduce labeling effort in 3 stages.
- Stage 1
Train the Encoder with all available data - Stage 2:
Fine-tune the SVM/Classifier layer with labeled data while freezing the weights for the encoder - Stage 3:
Using unlabeled data as input, the model with the weights frozen will output samples from least to highest informative/representative via the acquisition function. The top samples will be sent to a human annotator to be labeled.
The above stages are repeated until the total labeling budget finishes.
Reference: Reducing Label Effort: Self-Supervised meets Active Learning
