Reinforcement Learning
Domain: Data & Datasets
🎯 Motivation
Learning by trial and error. The model gets rewards for good actions and penalties for bad ones.
📋 Concrete Example
A game-playing AI: it tries moves, gets points (rewards) or loses points (penalties). Over millions of games, it learns the best strategies. AlphaGo learned this way.
🔗 Analogy
Training a dog - correct behavior gets a treat (reward), bad behavior gets nothing (penalty). The dog learns through trial and error. RL trains AI the same way.