A data scientist observes that a model performs well on training data but poorly on new data. They decide to apply L2 regularization. In the context of the bias-variance tradeoff, what effect does increasing the L2 penalty (lambda) have?

A is correct because L2 regularization (Ridge) penalizes large coefficient values, shrinking them toward zero. This introduces bias (the model is constrained from fitting the training data perfectly) but reduces variance (the model is less sensitive to noise in the training set), which addresses overfitting. B is wrong — decreasing bias while increasing variance would worsen overfitting. C is wrong because bias and variance respond oppositely to regularization strength. D is wrong because there is no free-lunch way to simultaneously decrease both.