Microsoft Azure AI Fundamentals (AI-900)Describe fundamental principles of machine learning on AzureHard
A team is developing an AI system for a self-driving car. The system needs to learn how to navigate complex urban environments by making decisions (e.g., accelerate, brake, turn) based on sensor inputs, receiving rewards for safe and efficient driving, and penalties for mistakes. Which machine learning paradigm is most appropriate for this type of problem?
- ATransfer Learning
- BUnsupervised Learning
- CReinforcement Learning
- DSupervised Learning
Show answer & explanationAnswer & explanation
Correct answer: C. Reinforcement Learning
Reinforcement learning is ideal for scenarios where an agent learns through trial and error by interacting with an environment, receiving rewards for desired actions and penalties for undesired ones. This perfectly describes the process of training a self-driving car to navigate dynamically.
Why the other options are wrong
- A. Transfer learning involves reusing a pre-trained model for a new, related task, which is a technique, not a fundamental learning paradigm for this problem.
- B. Unsupervised learning focuses on finding patterns in unlabeled data, not on sequential decision-making in an environment.
- D. Supervised learning requires labeled input-output pairs, which are hard to define for every possible driving scenario.
Reinforcement Learning
A machine learning paradigm where an agent learns to make optimal decisions by interacting with an environment, receiving rewards for desired actions and penalties for undesired ones, aiming to maximize cumulative reward.
- Involves an agent, environment, actions, states, and rewards.
- Learns through trial and error.
- Often used for control problems, robotics, game playing, autonomous systems.
Memory trick: Reinforcement 'Rewards' the agent for 'Right' driving decisions.