// ◇ Classical Machine Learning
Regularized Regression (Lasso / Ridge)
Penalize linear-model weights to curb overfitting and select features.
正则化回归（Lasso / Ridge） aka Ridge / L1/L2 Regularization / Elastic Net / 岭回归 1996 Foundational core
IN PLAIN WORDS
Besides small error, also ask for small weights. An L2 penalty (Ridge) shrinks all weights; an L1 penalty (Lasso) drives unimportant ones to exactly zero, selecting features automatically. The penalty strength is a hyperparameter – the clearest entry point to regularization.
Used in: Feature selection on high-dimensional data, sparse regression, gene-expression analysis
01Original paper
Regression Shrinkage and Selection Via the Lasso
Robert Tibshirani · Journal of the Royal Statistical Society Series B · 1996
DOI ↗ ✓ Verified · Crossref API doi:10.1111/j.2517-6161.1996.tb02080.x
02Where it sits
Learn the left column first; the right column builds on this one.
Learn first
Then you can learn
- Nothing in the atlas builds on it yet
Related:Logistic RegressionSupport Vector Machine
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
04Where the course teaches it
For AI / data
- Verification: verified, checked 2026-10-22.
- JSON: /api/learn/algorithms.json (id:
lasso)