Classical Machine Learning

Separate the inner and outer groups

An outer point lies on the other side of the centre

The inner and outer point groups need a coordinate that measures distance from the centre. Botie will make that feature visible before using an implicit kernel comparison.

BOTIE

The changed point is [-2, 0], while the baseline inner point is [0, 1]. Which description should determine their groups?

The inputs we are comparing

Point coordinates

Example 1

Example 4

Result for example 1

Choose an explanation to test

Squared distance from the centreWhether the first coordinate is positiveWhether either coordinate is zero

A point exactly on the lifted threshold

This check uses the pieces from “Separate groups with the radius feature”.

BOTIE

Both original coordinates equal one. Predict the class under the workshop's rule that the squared-radius feature must be strictly greater than two.

Point coordinates: 01
11

What result should your machine produce for this case?

Use a number, text, or an array matching the requested result. For a matrix, put each row in brackets. Saving your prediction runs your current construction on these inputs and checks the earlier examples too.

Build a construction in the workspace above first. This case will test those pieces.

What has become simple after changing the representation, and what distinguishes this explicit construction from an actual kernel trick?

Your notes for later. These stay with this journey.

Changing a representation can simplify a decision, but a visible feature map and an implicit kernel computation are different mechanisms.

This explicitly constructs a feature. The kernel trick computes feature-space comparisons without explicitly constructing every feature.

Work through the Python challenges →

Continue the courseKernel Ridge Regression →

← Density-Based Clustering