Interview questions

Frequently asked questions with concise model answers. Try answering aloud before expanding.

What is dropout? Medium

A regularisation technique that randomly zeros a fraction of neurons during training, forcing the network to learn redundant representations and reducing overfitting.

Why use ReLU over sigmoid in hidden layers? Medium

ReLU doesn't saturate for positive inputs, mitigating vanishing gradients, and is cheap to compute — leading to faster training in deep networks.