Module 2 — Core ML Theory·Lesson 5 of 22

2.2 Regularization

Interview-ready field notes

The 20-second answer

Regularization deliberately limits or perturbs a model so it learns patterns that transfer beyond its training examples.

Core ideas to retain

  • L1 adds λw\lambda\sum|w| and can drive weights exactly to zero; L2 adds λw2\lambda\sum w^2 and smoothly shrinks them.
  • Dropout randomly removes activations during training, reducing brittle co-adaptation. Turn it off for inference.
  • Early stopping selects the checkpoint with the best validation score before the model memorizes noise.
  • Augmentation adds label-preserving variation, such as image crops or text paraphrases, to increase effective data diversity.

Interview / OA rule

Name the symptom before the fix. Regularization is for a generalization gap, not for a model that cannot fit the training set.

One good written resource

Google — Overfitting and Regularization — read this after the video when you want a clearer mental model, not more pages of notes.

Most asked

Interview questions to practise aloud

Each answer is the level of detail expected for a strong fundamentals round.

01

L1 vs L2 regularization?

L1 tends to produce sparse, exactly-zero weights; L2 shrinks all weights smoothly and is the usual weight-decay choice in neural nets.

02

Why does dropout help?

A unit cannot rely on a specific companion always being present, so features become less co-dependent. It behaves like averaging many thinned networks.

03

Is batch normalization regularization?

Its main job is optimization/stable activations; batch noise can add a mild regularizing effect, but it is not a substitute for validation-based regularization.