Definition
Discriminative Learning
Discriminative learning is a strategy in supervised learning that directly models the conditional probability or the decision boundary between classes. Formally, given an instance , the learner seeks to identify a mapping that distinguishes between labels without attempting to model the underlying distribution of the inputs .
Characteristics
Focus: Discriminative models prioritise the classification task directly, often yielding superior predictive performance when the number of samples is large.
Examples: Common discriminative algorithms include Logistic Regression, Support Vector Machines, and Neural Networks.