Society of Actuaries (SOA) PA Practice Exam

What defines supervised learning?

1. Learning from data without labels

2. Learning a mapping from input to output using example pairs

3. Learning from data where output is not known

4. Learning that focuses on clustering data

The correct answer is: Learning a mapping from input to output using example pairs

Supervised learning is defined as the process of learning a mapping from input to output using example pairs. In this context, it involves training a model on a labeled dataset, where each input is associated with a corresponding output (label). The goal is to enable the model to make predictions on new, unseen data by generalizing from the examples it learned during training. The emphasis on using example pairs highlights the importance of having both the features (inputs) and the labels (outputs) clearly defined. This enables the algorithm to adjust its parameters based on the relationship between the input data and the corresponding outputs, effectively learning the underlying patterns and associations in the data. Other concepts mentioned, such as learning from data without labels or where output is not known, relate more to unsupervised learning techniques. Similarly, focusing on clustering data also pertains to unsupervised learning, where the objective is to group similar data points without predefined labels. In contrast, the crux of supervised learning lies in its reliance on labeled data to train models for prediction tasks.