Short Answer

Why a location indicator can resolve mixed housing-price data

Question: Suppose a training set mixes home-sale records from two cities with different price levels and you add a city label to each feature vector. What is the theoretical effect on the target value y?

Sample answer: The city label conditions the model on the data source, so the same house features no longer point to one ambiguous price. For a fixed x, y becomes well-defined because the model can tell which city generated the example.

Key points:

  • The city label identifies which data source produced the example.
  • Once the source is identified, the mapping from x to y is no longer ambiguous.
  • The model can learn different relationships for different cities.

Rubric: The answer must state that the city label removes ambiguity by distinguishing the source so y is well-defined for a given x.

0

1

Updated 2026-08-12

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy

Machine Learning Yearning @ DeepLearning.AI

Related