Regarding Artificial Intelligence (AI), identify the correct statements using the codes given below : A. Machine learning is a subset of AI. B. Supervised learning involves training a computer system without labelled input data. C. Machines learn through trial and error, receiving rewards during reinforcement learning. Codes :

Correct — option (2), Only A and C.

A. True. Machine learning is a subset of artificial intelligence, focused on algorithms that learn the patterns of training data.

B. False. Supervised learning is defined by labelled data: the model trains on inputs paired with their correct outputs, the 'ground truth'. Training without labelled data describes unsupervised learning.

C. True. In reinforcement learning, a system learns by trial and error, and its objective is to maximise reward.

A and C are correct and B is not, so the code is Only A and C. B appears in every other code, so ruling it out decides the question.

The idea to remember: supervised learning needs labels, unsupervised learning works without them, and reinforcement learning learns from rewards.

Why the others are wrong

• (1)
  Only A and B — Statement A is right, but B is not. Supervised learning trains on labelled data — inputs paired with the correct answers — so 'without labelled input data' reverses its definition.

  This code also leaves out C, which is true: reinforcement learning works through trial and error, guided by rewards.

• (3)
  Only B and C — Statement C is right, but B is false and A is left out. Machine learning is a subset of AI, so A is true.

  Statement B describes training without labels, which is unsupervised learning, not supervised learning.

• (4)
  A, B and C all statements are correct. — A and C are correct, but B is not, so all three cannot be. Supervised learning is the kind that uses labelled data.

  Finding patterns such as groupings or correlations in unlabelled data is unsupervised learning.

Artificial intelligence is the broad field of making machines perform tasks that would otherwise need human intelligence. Machine learning is the part of AI in which systems learn patterns from data rather than following only hand-written rules.

Machine learning methods differ by the feedback used in training. Supervised learning trains on labelled examples, for tasks such as classification and regression. Unsupervised learning finds structure, such as groupings and correlations, in unlabelled data.

Reinforcement learning trains an agent by trial and error: it acts, receives rewards, and adjusts its behaviour to maximise reward over time.

Key facts

• Machine learning is a subset of artificial intelligence; deep learning is a subset of machine learning that uses multi-layered neural networks.
• Supervised learning trains models on labelled data, which provides the 'ground truth' linking inputs to outputs.
• Unsupervised learning finds patterns such as similarities, correlations and groupings in unlabelled data.
• Reinforcement learning trains through trial and error, with the objective of maximising reward.
• Semi-supervised learning uses both labelled and unlabelled data.

Common traps

• Supervised and unsupervised learning differ on one point: whether the training data are labelled. A statement that pairs 'supervised' with 'without labels' has swapped the two.
• Reinforcement learning does not learn from a labelled answer key; it learns from rewards for its own actions. Do not describe it as supervised learning.
• The nesting runs AI ⊃ machine learning ⊃ deep learning. A statement calling AI a subset of machine learning reverses it.

UnlockIAS compared this question with questions from other RAS Prelims papers and found none similar enough to link.

• practice — not a real PYQ

  Grouping customers into segments from unlabelled purchase data, with no predefined categories, is an example of –

  • (a)Supervised learning
  • (b)Unsupervised learning
  • (c)Reinforcement learning
  • (d)Rule-based expert system
  Answer(2) — Finding groupings in unlabelled data is unsupervised learning. Option (1) needs labelled examples, option (3) learns from rewards for actions, and option (4) follows hand-written rules instead of learning from data.
• practice — not a real PYQ

  Arrange the following from the broadest to the narrowest : A. Deep learning B. Artificial intelligence C. Machine learning Choose the correct option :

  • (a)B, C, A
  • (b)C, B, A
  • (c)B, A, C
  • (d)A, C, B
  Answer(1) — AI contains machine learning, which contains deep learning. Option (2) puts machine learning above AI, option (3) puts deep learning above machine learning, and option (4) reverses the whole order.