Learn by Source
University of Washington · on Coursera

Machine Learning: Regression

Beginner English SpecializationFREE
SkillsRegressionLinear regressionRidge and lassoModel assessmentMachine learningFeature selection

Is this course right for you?

Our take
The regression course of the University of Washington's ML Specialization, taught by Emily Fox and Carlos Guestrin around concrete case studies like predicting house prices.

Good for: Building solid conceptual understanding of regression, via case studies.

Skip if: You want current, job-ready tooling (scikit-learn/PyTorch) from the start.

The teaching of the concepts is genuinely strong — linear regression, ridge, lasso and how to assess a model — and the case-study framing keeps the ideas concrete rather than abstract. There's one significant caveat, though: the course dates from 2016 and works through the proprietary GraphLab/Turi Create library instead of today's standard scikit-learn or PyTorch.

So take it for understanding, not for job-ready code: the reasoning transfers cleanly, but you'll re-learn the tooling elsewhere. Free to audit on Coursera; a certificate needs a subscription (about $49/month) or Coursera Plus, with financial aid available. The statistical ideas themselves haven't aged (as of 2026).

Comparison · LBS

Compare alternatives for Machine Learning: Regression

Same topic, different options. We surface the trade-offs others hide so you can pick the course that actually fits your time, budget, and goals.
Coursera
Machine Learning: Regression
Price
Free
Audit free · Certificate available
Duration
—
Level
Beginner
Certificate
Specialization
Coursera
Machine Learning: Classification
Price
Free
Audit free · Certificate available
Duration
—
Level
Beginner
Certificate
Specialization
View course →
Google4.7(8,500)
Machine Learning Crash Course
Price
Free
Completely free, self-paced
Duration
15 hrs
Level
Beginner
Certificate
View course →
Coursera4.9(39,000)
Machine Learning Specialization
Price
Free
Audit free · Certificate on subscription
Duration
86 hrs
Level
Intermediate
Certificate
Specialization
View course →
Prices & availability can change — confirm on the provider's site. We're not affiliated with any single provider.

About this course

Machine Learning: Regression is the regression-focused course in the University of Washington's well-regarded ML Specialization. Through practical case studies (like predicting house prices), Emily Fox and Carlos Guestrin build up from simple linear regression to regularisation with ridge and lasso, feature selection, and the bias-variance trade-off — coding the ideas, not just describing them.

Browse all AI & ML courses Browse all Coursera courses

Instructor

EF
Emily Fox
Coursera instructor

Emily Fox and Carlos Guestrin are machine-learning researchers at the University of Washington whose lectures are often singled out as some of the clearest in any online ML course. The specialization takes a practical, case-study-driven approach.

Frequently asked questions

Partly — a caution here. It is part of the University of Washington's Machine Learning specialization, which is well-regarded conceptually but somewhat older, and historically used the GraphLab/Turi Create toolkit rather than the now-standard scikit-learn. The regression concepts it teaches are timeless, but the specific tooling is dated, so you may prefer to apply the ideas in scikit-learn and treat the course as concept-focused.

Yes to both, at a reasonable level. It involves programming assignments and assumes comfortable Python, plus a grasp of the maths behind regression — some linear algebra and calculus intuition. It is not a beginner's introduction to either. Arriving with Python and basic ML-maths grounding is the right preparation; without them, the material and assignments become a struggle.

For most people starting out today, Andrew Ng's Machine Learning specialization is the more current and widely recommended choice — updated tooling, huge community, and a gentle on-ramp. This UW course goes deep on regression specifically and is excellent conceptually, but its older toolkit makes it less practical as a first course. Take Ng's for a modern foundation; take this for focused depth on regression if you want it.

Yes. It is one course in the University of Washington's Machine Learning specialization, which covers regression, classification, clustering, and retrieval across its courses. You can take this regression course standalone for the focused material, but it is designed within that sequence, so the wider specialization builds around it if you continue.

Yes, you can audit it on Coursera and watch the lectures at no cost. The graded programming assignments and the certificate require a subscription, with financial aid available if you apply. Since the value is in the concepts and exercises, auditing gives you the teaching — though given the dated tooling, you may get more from applying the ideas in scikit-learn yourself.
Free
to audit
Enroll now