CAIC practice question 8 of 10

If humans are unlabeling the data and the machine is correctly labeling current or future data points, it’s ______.

Full set: 70 questions
Answer options
A: supervised learning
B: unsupervised learning
C: reinforcement learning
D: semi-reinforcement learning
E: semi-supervised learning Correct answer
Correct Answer:
E. semi-supervised learning
Explanation

Semi-supervised learning is a machine learning paradigm that utilizes a combination of a small amount of labeled data with a large amount of unlabeled data for training. The model learns the underlying structure from the unlabeled data and uses the labeled data to ground its predictions, effectively propagating labels to the unlabeled points. This approach is ideal when data labeling is expensive or time-consuming. The question, despite its awkward phrasing ("unlabeling the data"), describes a scenario where a machine learns to label data, which is characteristic of this method when a mix of labeled and unlabeled inputs are available.

Why Incorrect

A. Supervised learning requires a fully labeled dataset for training, which is not the case here.
B. Unsupervised learning works with entirely unlabeled data to find patterns, not to assign predefined labels.
C. Reinforcement learning involves an agent learning through rewards and penalties, which is a different paradigm.
D. Semi-reinforcement learning is not a standard, widely recognized term for this specific machine learning category.

References

1. Chapelle, O., Schölkopf, B., & Zien, A. (2006). Semi-Supervised Learning. MIT Press. In the Introduction (Chapter 1, p. 1), the book defines semi-supervised learning as the case where the learner is given a set of labeled examples and a set of unlabeled examples.

2. Zhu, X. (2005). Semi-Supervised Learning Literature Survey. University of Wisconsin-Madison, Computer Sciences Technical Report 1530. Section 1, "Introduction," defines semi-supervised learning as learning from a combination of labeled and unlabeled data.

3. Van Engelen, J. E., & Hoos, H. H. (2020). A survey on semi-supervised learning. Machine Learning, 109(2), 373-440. https://doi.org/10.1007/s10994-019-05855-6. The abstract clearly defines the paradigm as leveraging unlabeled data in addition to labeled data.

Scroll to Top