How models actually learn, from vanilla gradient descent to Adam
Sometimes parameters are not allowed to move anywhere. They must satisfy constraints: nonnegative weights, bounded norms, probabilities that must stay nonnegative and sum to 1 (a set called the probability simplex), fairness limits, safety limits, or physical feasibility.
🔒 This is a Pro lesson — the interactive figure, worked examples, quiz and practice open with Pro access.
▶ Constrained Optimization & Projections