Answer: Reinforcement learning

## Explanation This scenario describes **reinforcement learning** because: - **Reward/Penalty System**: The AI receives rewards for faster routes (positive reinforcement) and penalties for delays (negative reinforcement) - **Learning from Experience**: The system improves by learning from previous delivery outcomes - **Decision Optimization**: The goal is to optimize route decisions based on feedback **Why other options are incorrect**: - **A) Unsupervised learning**: Used for finding patterns in unlabeled data, not for reward-based optimization - **B) Self-supervised learning**: Involves creating labels from the data itself, not reward-based decision making - **D) Transfer learning**: Focuses on applying knowledge from one domain to another, not reward-based optimization Reinforcement learning is specifically designed for scenarios where an agent learns to make decisions by interacting with an environment and receiving rewards or penalties for its actions.

Author: Ritesh Yadav