Decision boundary

As explained in Machine Learning System Design

A decision boundary is the boundary in an input space that separates regions assigned different model decisions or predicted classes.

A decision boundary separates regions assigned different model decisions. Its margin matters: a narrow, non-robust boundary can make precision and recall change substantially when small noise is added. section 7.4

Because precision and recall depend on the selected threshold, evaluate boundary stability rather than relying on a single metric value. Poisson bootstrap, noise addition, and lower confidence bounds provide more reliable performance estimates. section 7.4

Testing boundary robustness

Compare precision and recall distributions for bootstrapped datasets with and without normally distributed noise. Large shifts, such as increased recall paired with decreased precision, indicate a narrow margin. section 7.4

Distribution of precision and recall with sample size of 200;000; every point is a bootstrapped original dataset, no noise added.

Inspect the bootstrapped precision and recall distribution without added noise, then compare it with the corresponding noisy distribution. section 7.4

How Machine Learning System Design covers it

How deep it goes

Explains robustness through metric distributions, Poisson bootstrap, noise addition, and confidence bounds without deriving boundary geometry.

Examples

Uses precision and recall distributions from bootstrapped datasets with and without added noise.

What you can do after

Diagnose boundary fragility by comparing metric shifts and applying robust confidence-bound estimates.

How it connects

Connects decision boundaries with classification thresholds, precision, recall, specificity, Poisson bootstrap, and confidence intervals.

liveBookmanning.com

Read Machine Learning System Design on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Math and Architectures of Deep LearningA decision boundary that is a straight line or hyperplane, unable to separate the XOR points in this example.
• 

  liveBookmanning.com
  Machine Learning Algorithms in DepthA decision boundary is a surface in feature space that separates regions assigned to different predicted classes.
• 

  liveBookmanning.com
  Outlier Detection in PythonA decision boundary is a surface in feature space that separates regions assigned different model decisions or classes.

Compare how 4 books explain decision boundary →

Related concepts

• hyperplaneincludes · A hyperplane is a flat affine subspace with dimension one less than its ambient space, capable of dividing that space into regions.
• support vectorused by · A support vector is a training example that contributes to determining a support vector machine’s learned prediction function or decision boundary.
• linear discriminant analysisproduced by · Linear discriminant analysis is a supervised method that projects data to separate labeled classes and supports linear classification.
• classifierequivalent to · A classifier is a predictive computational model that assigns inputs to one or more discrete classes or categories.
• discriminative modelused by · A discriminative model predicts labels or other outcomes from inputs by modeling conditional relationships or decision boundaries rather than the full data-generating distribution.

This explainer was generated by Manning's knowledge graph from the text of Machine Learning System Design. The section links point to the parts of the book it draws on.