Data Scholar
ArticlesProjectsLearningAbout
Sign inSubscribe
Machine Learning Foundations

3 modules · 18 topics

  1. 1What machine learning is (and is not)
    • Overview
    • Learning from data versus writing rules
    • Supervised and unsupervised learning, and the task families
    • Data, features, labels, and examples
    • Agentic, not academic: what to learn and why
    • Framing a real problem as an ML project
    • Practice: name the task and frame a project
  2. 2The ML workflow and the generalization problem
    • Overview
    • The end-to-end machine learning workflow
    • Generalization: train versus unseen
    • Overfitting, underfitting, and bias-variance
    • Train, validation, and test, and cross-validation
    • Baselines and choosing the metric first
    • Practice: diagnose the fit
  3. 3The model zoo (algorithms as intuition)
    • Overview
    • Linear and logistic regression
    • k-nearest neighbors and the idea of similarity
    • Decision trees
    • Ensembles and gradient boosting
    • Choosing a model, and why tuning beats choosing
    • Practice: pick the model
← Machine Learning Foundations
Course contents
Machine Learning Foundations

3 modules · 18 topics

  1. 1What machine learning is (and is not)
    • Overview
    • Learning from data versus writing rules
    • Supervised and unsupervised learning, and the task families
    • Data, features, labels, and examples
    • Agentic, not academic: what to learn and why
    • Framing a real problem as an ML project
    • Practice: name the task and frame a project
  2. 2The ML workflow and the generalization problem
    • Overview
    • The end-to-end machine learning workflow
    • Generalization: train versus unseen
    • Overfitting, underfitting, and bias-variance
    • Train, validation, and test, and cross-validation
    • Baselines and choosing the metric first
    • Practice: diagnose the fit
  3. 3The model zoo (algorithms as intuition)
    • Overview
    • Linear and logistic regression
    • k-nearest neighbors and the idea of similarity
    • Decision trees
    • Ensembles and gradient boosting
    • Choosing a model, and why tuning beats choosing
    • Practice: pick the model

Topic · What machine learning is (and is not)

Learning from data versus writing rules


Sign in to continue

Course content is free with an account. Sign in to read this and keep your progress.

Sign inCreate account
Data Scholar

Data Scholar is where data and AI get learned properly, from first principles to production: articles, end-to-end build write-ups, free courses, and open-source projects for data analysts, data engineers, and AI and GenAI engineers.

Explore

  • Articles
  • Projects
  • Learning
  • About

Get updates

New articles, projects, and courses, straight to your inbox.

Subscribe

© 2026 Data Scholar. Always free.

PrivacyTerms of Use