Interview Questions and Answers

Supervised learning involves training a model on a labeled dataset, while unsupervised learning deals with unlabeled data where the algorithm tries to find patterns or relationships on its own.

Cross-validation is a technique used to assess the performance of a model by dividing the dataset into multiple subsets, training the model on some, and testing on others. It helps to obtain a more reliable estimate of a model's performance.

Overfitting occurs when a model learns the training data too well, capturing noise and producing poor generalization on new data. Regularization techniques, cross-validation, and increasing training data are common methods to prevent overfitting.

A decision tree is a tree-like model where each node represents a decision based on a feature, and each branch represents an outcome of that decision. It is used for both classification and regression tasks.

Batch gradient descent updates the model parameters using the entire dataset, while stochastic gradient descent updates the parameters using one randomly selected data point at a time. Mini-batch gradient descent is a compromise, using a small subset of the data for each update.

KNN is a simple, instance-based learning algorithm used for classification and regression. It classifies a new data point based on the majority class of its k-nearest neighbors in the feature space.

The Receiver Operating Characteristic (ROC) curve is a graphical representation of a binary classification model's performance across different thresholds. It plots the true positive rate against the false positive rate, helping to assess the trade-off between sensitivity and specificity.

Dropout is a regularization technique used in neural networks to randomly deactivate some neurons during training. It helps prevent overfitting and encourages the network to learn more robust features.

Precision is the ratio of correctly predicted positive observations to the total predicted positives, while recall is the ratio of correctly predicted positive observations to the total actual positives. Precision emphasizes the accuracy of positive predictions, while recall focuses on capturing all positive instances.

Cross-entropy loss, or log loss, measures the performance of a classification model whose output is a probability value between 0 and 1. It penalizes models that are confidently wrong and is a common choice for binary and multiclass classification problems.

Precision is the ratio of true positives to the sum of true positives and false positives, while the F1 score is the harmonic mean of precision and recall. F1 score provides a balance between precision and recall, giving equal weight to both metrics.

Feature importance measures the contribution of each feature to the predictive performance of a model. It helps identify the most influential features in making predictions and is often used for feature selection and model interpretation.

Bias refers to the error introduced by approximating a real-world problem with a simplified model. Variance is the amount by which the model's prediction would change if it were estimated using a different training dataset. The bias-variance tradeoff aims to balance these two sources of error.

Ensemble learning combines the predictions of multiple models to improve overall performance. Common ensemble techniques include bagging, boosting, and stacking. The idea is that the combination of diverse models can provide better results than individual models.