AI and machine learning interview questions

fit learns parameters from data, such as column means for a scaler.

A neural net stacks layers of weighted sums and nonlinear activations.

Without nonlinear activations, stacked layers collapse into one linear map.

The loss scores how wrong the predictions are, such as cross-entropy for classification.

Regularization penalizes complexity so the model prefers simpler weights or smaller trees.

L2, also called ridge, shrinks weights toward zero but rarely makes them exact zeros.

Dropout randomly turns off units during training so the net cannot rely on any single path.

Cross-validation estimates generalization by training and scoring on several splits of the same data.

You cut the data into k folds, train on k minus one, and score the held-out fold.

Leave-one-out trains almost n models, which is expensive.

Semi-supervised learning mixes a small labeled set with a large unlabeled set.

Self-supervised learning invents a pretext task from the data itself, such as predicting a masked token or an image rotation.

A discriminative model, such as logistic regression, models P(y|x) and is usually stronger for pure prediction.

If training and validation error are both high and close, you mostly have bias or an unsolvable task.

More independent training data usually lowers variance because the fit is less hostage to any one sample.

K equal to 1 is a high-variance, low-bias classifier that can zigzag around every point.

Stronger L2 or smaller trees increase bias by shrinking or simplifying the fit.

Early stopping tracks a validation metric and freezes weights when that metric stops improving.

Extra columns raise capacity and the chance of spurious correlations in a finite sample.

Interpolation means the model can fit the training points, even perfectly.