The 60-second answer
Logistic regression models log-odds as a linear function of features and maps them to probabilities with the sigmoid. Training minimizes binary cross-entropy / negative log likelihood; multiclass versions commonly use softmax cross-entropy.
Build the answer in this order
Logistic regression models log-odds as a linear function of features and maps them to probabilities with the sigmoid.
Training minimizes binary cross-entropy / negative log likelihood; multiclass versions commonly use softmax cross-entropy.
Regularization controls variance and can induce sparsity; standardized features make penalty strength easier to interpret.
Evaluate discrimination and calibration separately, then choose thresholds based on error costs.
A useful interview mental model
This is the shape of a strong answer—not a script to memorize.
Senior-level signal
- Perfect separation can drive unregularized maximum-likelihood coefficients toward infinity.
- For production decisions, monitor calibration and base-rate shift, not only AUC.
What the interviewer is really testing
Likely follow-up questions
Common weak-answer patterns
- Reciting a definition without mechanism or assumptions.
- Claiming one technique is always better without a data regime.
- Stopping before failure modes, validation, or deployment implications.