Q11. What is regularization and how do L1 and L2 differ?
Regularization adds a penalty on model complexity to the loss, discouraging large weights and reducing overfitting. L2 (ridge) adds the sum of squared weights. It shrinks...
Technical Question Bank · Peer-Reviewed
Job Interview Questions and Answers Directory
Search and filter frequently asked interview questions across technical, programming, and behavioral domains. Tailor your interview preparation by difficulty for freshers or experienced developers (18 available).
Active:
Tag: learning ×
Q12. Explain how gradient descent optimizes a model.
Gradient descent minimizes a loss function by repeatedly moving parameters in the direction that decreases the loss. The gradient is the vector of partial derivatives, an...
Q13. How do you evaluate a regression model?
Choose metrics that match the problem. - MAE: mean absolute error, in the target's units, robust to outliers. - MSE and RMSE: penalize large errors more; RMSE is in targe...
Q14. Compare random forests and gradient boosting.
Both are tree ensembles, differing in how trees are combined. Random forest builds many deep, independent trees on bootstrap samples with random feature subsets, then ave...
Q15. What is data leakage in machine learning?
Data leakage is when information unavailable at prediction time seeps into training, producing optimistic validation scores that collapse in production. Common sources: -...
Q16. What is MLOps and why does monitoring matter?
MLOps applies DevOps practices to machine learning: versioning data, code and models; automating training and deployment; and monitoring models in production. Key compone...
Q17. Explain ROC-AUC and when it can be misleading.
ROC-AUC is the probability that a randomly chosen positive example is ranked above a randomly chosen negative one. It summarizes performance across all thresholds, and 0....
Q18. How would you design a recommendation system end to end?
Start with the objective and data: implicit signals such as views, clicks and purchases, plus explicit ratings and item or user features. A common two-stage design: - Can...