Reinforcement learning¶
Reinforcement learning (RL) trains an agent to act in an environment to maximize cumulative reward — no labeled dataset, just experience. Where supervised learning (§13, §18) hands the model a fixed table of \((x, y)\) pairs and asks it to fit \(P(y \mid x)\), RL hands the model nothing but a stream of consequences: it acts, the world reacts, and the only signal telling it whether that was a good idea is a scalar reward that may arrive many steps later. …
🔒 La suite est en accès freemium — lecture complète 100 % gratuite
Tu lis ici l'aperçu libre. Le reste du chapitre (code, schémas, maths, exercices) fait partie du livre complet : crée un compte gratuit (30 secondes, aucun paiement) pour tout lire.