TK Study
HomeMemoriseFlashcardsNotesMaster answersPast questionsMultiple choiceMock examPractice & marker

Multiple choice › Machine learning training

Machine learning training

1. Machine learning training

A model is trained on photos that humans have already labelled 'cat' or 'dog'. This is:

  1. Semi-supervised learning
  2. Supervised learning
  3. Reinforcement learning
  4. Unsupervised learning
Answer

B

Supervised learning uses labelled examples (inputs with known outputs).

2. Machine learning training

An algorithm groups customers into segments using purchase data that has no labels. This is:

  1. Logistic regression
  2. Reinforcement learning
  3. Unsupervised learning
  4. Supervised learning
Answer

C

Unsupervised learning finds structure (clusters) in unlabelled data.

3. Machine learning training

A robot learns to walk by receiving a reward for each step forward and a penalty for falling. This is:

  1. Reinforcement learning
  2. Unsupervised learning
  3. Supervised learning
  4. Semi-supervised learning
Answer

A

Reinforcement learning: an agent learns by trial and error from rewards and penalties.

4. Machine learning training

A small set of labelled images and a very large set of unlabelled images are used together to train a model. This is:

  1. Pure unsupervised learning
  2. Reinforcement learning
  3. Semi-supervised learning
  4. Rule-based programming
Answer

C

Semi-supervised learning mixes a little labelled data with a lot of unlabelled data.