Which statement correctly defines supervised, unsupervised, and reinforcement learning with an example for each?

Which statement correctly defines supervised, unsupervised, and reinforcement learning with an example for each?

• A. Supervised learns by interacting with environment; unsupervised uses labeled data; reinforcement uses static datasets.
• B. Supervised learns from labeled data; unsupervised finds patterns; reinforcement learns by interacting with environment.
• C. Supervised uses unlabeled data; unsupervised uses labeled data; reinforcement uses static datasets.
• D. Supervised is always faster; unsupervised is always slower; reinforcement is not learning.

Answer: (A)

If you understand how each learning approach gets feedback and data, you’ll see why this description fits best. In supervised learning, the model is trained on data that comes with correct answers—labels. The goal is to learn a mapping from inputs to those labels, so you can predict the label for new inputs. A classic example is image classification: you train on labeled images (for instance, “cat” or “dog”) so the model can identify new images correctly. In unsupervised learning, there are no labels. The aim is to uncover structure, patterns, or groupings in the data. Examples include clustering customers into segments based on behavior, or reducing dimensionality to reveal relationships in the data. In reinforcement learning, the learning agent interacts with an environment, takes actions, and receives rewards or punishments that guide future behavior. Over time, the agent learns a policy that maximizes cumulative reward. A typical example is a robot learning to navigate a maze or a game-playing agent improving through trial and error. So the statement that supervised learning uses labeled data, unsupervised learning finds patterns in unlabeled data, and reinforcement learning learns by interacting with the environment captures the correct distinctions.

If you understand how each learning approach gets feedback and data, you’ll see why this description fits best. In supervised learning, the model is trained on data that comes with correct answers—labels. The goal is to learn a mapping from inputs to those labels, so you can predict the label for new inputs. A classic example is image classification: you train on labeled images (for instance, “cat” or “dog”) so the model can identify new images correctly.

In unsupervised learning, there are no labels. The aim is to uncover structure, patterns, or groupings in the data. Examples include clustering customers into segments based on behavior, or reducing dimensionality to reveal relationships in the data.

In reinforcement learning, the learning agent interacts with an environment, takes actions, and receives rewards or punishments that guide future behavior. Over time, the agent learns a policy that maximizes cumulative reward. A typical example is a robot learning to navigate a maze or a game-playing agent improving through trial and error.

So the statement that supervised learning uses labeled data, unsupervised learning finds patterns in unlabeled data, and reinforcement learning learns by interacting with the environment captures the correct distinctions.