dc dotCreds
Reference guide

Stanford Machine Learning Course Notes

Study Stanford Machine Learning section notes, then jump straight into the guided course or related practice questions without losing your place.

Continue Course Start Practice
Checking access

Checking Pro access...

Looking for your active Pro access before showing Course Notes. This usually takes just a moment.

Course Notes preview

Unlock Pro for the full per-exam reference guide.

Preview one piece from each section. Pro includes every Course Notes section, summary, key point, common mistake, exam tip, related-question review, and PDF export.

Includes full Course Mode and Course Notes.

Section 1 Foundations: Supervised Learning Preview
More in this section
  • Full summary in Pro version
  • 9 more key points in Pro version
  • 28 more related questions in Pro version

Summary

Supervised learning trains a model from examples that include both inputs and correct outputs. The model learns a mapping from features to a target, then uses that learned relationship to predict outputs for new examples.

Key Points

  • Supervised learning trains a model from examples that include both inputs and correct outputs. The model learns a mapping from features to a target, then uses that learned relationship to predict outputs for new examples.

Common Mistakes

No common mistakes are available for this section yet.

Exam Tips

No exam tips are available for this section yet.

Section 2 Model Selection & Evaluation Preview
More in this section
  • Full summary in Pro version
  • 9 more key points in Pro version
  • 32 more related questions in Pro version

Summary

Model evaluation separates fitting from honest measurement. The training set teaches the model, the validation set guides model and hyperparameter choices, and the test set estimates performance after choices are finalized.

Key Points

  • Model evaluation separates fitting from honest measurement. The training set teaches the model, the validation set guides model and hyperparameter choices, and the test set estimates performance after choices are finalized.

Common Mistakes

No common mistakes are available for this section yet.

Exam Tips

No exam tips are available for this section yet.

Section 3 Unsupervised Learning Techniques Preview
More in this section
  • Full summary in Pro version
  • 9 more key points in Pro version
  • 27 more related questions in Pro version

Summary

Unsupervised learning looks for structure in data without labeled target values. Instead of learning from correct answers, the algorithm groups examples, reduces dimensions, or estimates what normal behavior looks like.

Key Points

  • Unsupervised learning looks for structure in data without labeled target values. Instead of learning from correct answers, the algorithm groups examples, reduces dimensions, or estimates what normal behavior looks like.

Common Mistakes

No common mistakes are available for this section yet.

Exam Tips

No exam tips are available for this section yet.

Section 4 Recommender Systems Preview
More in this section
  • Full summary in Pro version
  • 9 more key points in Pro version
  • 11 more related questions in Pro version

Summary

Recommender systems predict which items a user may prefer, such as movies, products, articles, or courses. The central problem is ranking a large set of possible items so the most useful choices appear near the top.

Key Points

  • Recommender systems predict which items a user may prefer, such as movies, products, articles, or courses. The central problem is ranking a large set of possible items so the most useful choices appear near the top.

Common Mistakes

No common mistakes are available for this section yet.

Exam Tips

No exam tips are available for this section yet.

Section 5 Advanced Models: Trees & Recommenders Preview
More in this section
  • Full summary in Pro version
  • 9 more key points in Pro version
  • 12 more related questions in Pro version

Summary

Decision trees make predictions by asking a sequence of feature-based questions. Each split partitions the data, and each leaf gives a prediction such as a class label, probability, or numeric value.

Key Points

  • Decision trees make predictions by asking a sequence of feature-based questions. Each split partitions the data, and each leaf gives a prediction such as a class label, probability, or numeric value.

Common Mistakes

No common mistakes are available for this section yet.

Exam Tips

No exam tips are available for this section yet.

Section 6 Deep Learning Essentials Preview
More in this section
  • Full summary in Pro version
  • 9 more key points in Pro version
  • 17 more related questions in Pro version

Summary

A neural network is a composition of layers that transform inputs into predictions. Each layer applies weights, biases, and activation functions, allowing the model to learn nonlinear relationships.

Key Points

  • A neural network is a composition of layers that transform inputs into predictions. Each layer applies weights, biases, and activation functions, allowing the model to learn nonlinear relationships.

Common Mistakes

No common mistakes are available for this section yet.

Exam Tips

No exam tips are available for this section yet.

Section 7 Workflow & Pipelines Preview
More in this section
  • Full summary in Pro version
  • 9 more key points in Pro version
  • 14 more related questions in Pro version

Summary

A practical ML workflow starts with problem framing. The first decision is whether the task is regression, classification, clustering, anomaly detection, recommendation, or another formulation that matches the business question and available labels.

Key Points

  • A practical ML workflow starts with problem framing. The first decision is whether the task is regression, classification, clustering, anomaly detection, recommendation, or another formulation that matches the business question and available labels.

Common Mistakes

No common mistakes are available for this section yet.

Exam Tips

No exam tips are available for this section yet.

Section 8 Specialization Overview Preview
More in this section
  • Full summary in Pro version
  • 4 more key points in Pro version
  • 1 more related question in Pro version

Summary

The specialization spans the core supervised, unsupervised, tree-based, recommender, neural-network, and workflow ideas needed to reason about common machine learning problems.

Key Points

  • The specialization spans the core supervised, unsupervised, tree-based, recommender, neural-network, and workflow ideas needed to reason about common machine learning problems.

Common Mistakes

No common mistakes are available for this section yet.

Exam Tips

No exam tips are available for this section yet.