weakly supervised learning

Weakly supervised learning is a machine learning technique where a model is trained using partially labeled or less precise data instead of fully annotated data, allowing the model to learn with reduced supervision.

Requires login.