When would you use Group Lasso instead of standard Lasso?
hardAnswer
- Group Lasso applies an L2 penalty within groups of coefficients and an L1 penalty across groups: entire groups get selected or dropped together.
- Use it when features come in natural groups — one-hot dummies of a categorical variable, spline basis functions, features from the same sensor.
- Standard Lasso would drop some dummies of a category, which is semantically odd; Group Lasso keeps or drops the whole category.
Check yourself — multiple choice
- Group Lasso is faster than Lasso for large datasets
- It selects entire groups of features together (e.g., all one-hot dummies of a category)
- It replaces cross-validation
- It only works for regression trees
Group Lasso: entire groups in or out — useful for dummies, splines, sensor groups.
#regularization#feature-selection#features
Practise Supervised Learning
215 interview questions in this topic.
Related questions
- How do you handle strong multicollinearity in a linear model?
- What's the main risk of polynomial regression?
- Why does L1 (Lasso) produce sparse coefficients but L2 (Ridge) does not?
- What is Weight of Evidence (WoE) encoding?
- Filter, wrapper, and embedded feature selection — how do they differ?
- What is stability selection?