← All tutorials
Skill Zuno · skillzunos.com · Python ML
Bias & Variance tutorial
Learn Bias & Variance on Skill Zuno. This Bias & Variance tutorial covers New, Easy, Medium, Hard, and Diff levels so you can study practical Bias & Variance problems on skillzunos.com.
Open Bias & Variance on Skill ZunoAlso see the Bias & Variance interview questions.
The bias-variance trade-off explains why models make errors on new data. Bias is error from overly simple assumptions, which causes underfitting; variance is error from sensitivity to the particular training data, which causes overfitting. Interviews ask you to decompose prediction error, diagnose high bias or high variance from learning curves, and pick the right fix.
Bias & Variance topics covered
- Bias, variance and irreducible error — How expected prediction error splits into three parts.
- Model complexity — Why error on new data falls and then rises as models become more flexible.
- Learning curves — Reading training and validation scores against training set size to diagnose problems.
- Fixing high bias — More features, a more flexible model or less regularisation.
- Fixing high variance — More data, regularisation, simpler models, feature selection or bagging.
- Bias and variance in common algorithms — Why linear models lean towards bias and deep trees lean towards variance.
What you can learn
- Bias & Variance tutorial for beginners and experienced developers
- Level-based practice: New, Easy, Medium, Hard, and Diff
- Python ML topics alongside Bias & Variance on skillzunos.com