How does regularization affect bias and variance?

Updated May 15, 2026

Short answer

Regularization increases bias slightly but significantly reduces variance.

Deep explanation

Regularization techniques like L1 (Lasso) and L2 (Ridge) add penalties to model complexity. This discourages overfitting by shrinking coefficients or removing irrelevant features. As a result, variance decreases because the model becomes less sensitive to noise, but bias increases because flexibility is reduced.

Real-world example

Ridge regression stabilizes predictions in financial forecasting models with many correlated features.

Common mistakes

Assuming regularization always improves accuracy.

Follow-up questions

What is the difference between L1 and L2 regularization?
Can regularization cause underfitting?

More Bias & Variance interview questions

View all →

How does model initialization strategy in large neural networks affect bias and variance during training?senior
How does distributed model evaluation architecture affect bias and variance estimation reliability?senior
How does asynchronous feature pipeline updates impact bias and variance in ML systems?senior
How does dynamic model selection at inference time influence bias and variance in large-scale systems?senior
How does real-time model rollback architecture affect bias and variance in production ML systems?senior
How does distributed inference caching consistency affect bias and variance in global ML systems?senior
How does model compression pipeline design influence bias and variance in edge ML systems?senior
How does real-time feature computation latency affect bias and variance in streaming ML systems?senior