FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
{Consistency, Confidence, Pseudo-labeling}
Embedded Files
0) Motivation, Object and Related works:
Motivation:
Semi-supervised learning (SSL) provides an effective means of leveraging unlabeled data to improve a model’s performance. This domain has seen fast progress
recently, at the cost of requiring more complex methods.
Objectives: FixMatch
An algorithm that is a significant simplification of existing SSL methods.
FixMatch first generates pseudo-labels using the model’s predictions on weakly-augmented unlabeled images. For a given image, the pseudo-label is only retained if the model produces a high-confidence prediction.
The model is then trained to predict the pseudo-label when fed a strongly-augmented version of the same image.
Despite its simplicity, we show that FixMatch achieves state-of-the-art performance across a variety of standard semi-supervised learning benchmarks
94.93% accuracy on CIFAR-10 with 250 labels
88.61% accuracy with 40 – just 4 labels per class.
Global Framework:
Related Works:
1. Consistency Regularization:
2. SSL Comparison
References:
Page updated
Google Sites
Report abuse