Self-Supervised Learning uses the data itself to generate labels or training signals, eliminating the need for manual annotation. Tasks like predicting masked words in text or filling missing patches in images are common. Large-scale self-supervised learning has led to powerful pretrained models.