A) Does not require a loss function.
B) Is guaranteed to converge faster than any other method.
C) Is only applicable to linear models.
D) Offers a balance between the efficiency of batch GD and the robustness of SGD.

A) Using only unsupervised learning techniques.
B) Training a model from scratch on every new problem.
C) Taking a model pre-trained on a large dataset (e.g., ImageNet) and fine-tuning it for a new, specific task with a smaller dataset.
D) Forgetting everything a model has learned.

A) Reinforcement learning.
B) Supervised classification of images.
C) Predicting continuous values in a regression task.
D) Unsupervised learning tasks like dimensionality reduction and data denoising.b

A) An encoder that compresses the input and a decoder that reconstructs the input from the compression.
B) Only a single layer of perceptrons.
C) A single output neuron with a linear activation.
D) A convolutional layer followed by an RNN layer.

A) The harmonic mean of precision and recall.
B) The proportion of positive identifications that were actually correct.
C) The proportion of total predictions that were correct.
D) The proportion of actual positives that were identified correctly.

A) You need a single metric that combines precision and recall.
B) The cost of false negatives is high (e.g., in disease screening, where you don't want to miss a sick patient).
C) The cost of false positives is high (e.g., in spam detection, where you don't want to flag legitimate emails as spam).
D) You are evaluating a regression model.

A) You need a single metric that combines precision and recall.
B) You are evaluating a clustering model.
C) The cost of false positives is high (e.g., in spam detection).
D) The cost of false negatives is high (e.g., in disease screening, where you don't want to miss a sick patient).

A) The difference between precision and recall.
B) A metric used exclusively for regression.
C) The arithmetic mean of precision and recall.
D) The harmonic mean of precision and recall, providing a single score that balances both concerns.

A) The average of the squares of the errors between predicted and actual values.
B) The variance of the input features.
C) The accuracy of a classification model.
D) The total number of misclassified instances.

A) The clustering quality of a K-means algorithm.
B) The architecture of a neural network.
C) The performance of a binary classification model at various classification thresholds.
D) The loss of a regression model over time.

A) 0.0.
B) 1.0.
C) -1.0.
D) 0.5.

A) Replace the need for a separate test set.
B) Obtain a more robust estimate of model performance by training and evaluating the model K times on different splits of the data.
C) Visualize high-dimensional data.
D) Increase the size of the training dataset.

A) A single, pre-defined rule.
B) The majority vote among its K closest neighbors in the feature space.
C) A random selection from the training set.
D) The output of a linear function.u

A) Is the learning rate for the algorithm.
B) Is the number of features in the dataset.
C) Is always set to 1 for the best performance.
D) Controls the model's flexibility. A small K can lead to overfitting, while a large K can lead to underfitting.

A) Classifying data using a decision boundary.
B) Finding new, uncorrelated dimensions (principal components) that capture the maximum variance in the data.u
C) Clustering data into K groups.
D) Predicting a target variable using linear combinations of features.

A) Is perpendicular to all other components.
B) Captures the least possible variance in the data.
C) Captures the greatest possible variance in the data.
D) Is randomly oriented.

A) The data is perfectly classified into known labels.
B) The between-cluster variance is minimized.
C) The data is projected onto a single dimension.
D) The within-cluster variance is minimized.

A) Determine the learning rate for gradient descent.
B) Help choose the optimal number of clusters K by looking for a "bend" in the plot of within-cluster variance.
C) Evaluate the accuracy of a classification model.
D) Initialize the cluster centroids.

A) Always have the lowest possible accuracy.
B) Do not use probability in their predictions.
C) Are very simple and cannot handle complex data.
D) Make a strong (naive) assumption that all features are conditionally independent given the class label.

A) Dimensionality reduction technique.
B) Regression algorithm for predicting continuous values.
C) Clustering algorithm for grouping unlabeled data.
D) Classification algorithm that models the probability of a binary outcome using a logistic function.

A) Distance to the decision boundary.
B) Exact value of the target variable.
C) Probability that the input belongs to a particular class.
D) Number of features in the input.

A) Support Vector Machines.
B) Linear Regression models.
C) K-NN models.
D) Decision Trees to reduce overfitting and improve generalization.

A) Increase the speed of a single decision tree.
B) Reduce bias by making trees more complex.
C) Perform feature extraction like PCA.
D) Reduce variance by training individual trees on random subsets of the data and averaging their results

A) Do not require any parameter tuning.
B) Are exclusively used for unsupervised learning.
C) Build models sequentially, where each new model corrects the errors of the previous ones.
D) Build all models independently and average them.

A) The evaluation of a model's final performance.
B) The process of deleting all features from a dataset.
C) The process of using domain knowledge to create new input features that make machine learning algorithms work better.
D) The automatic learning of features by a deep neural network.

A) Normalize continuous numerical features.
B) Convert categorical variables into a binary (0/1) format that can be provided to ML algorithms.
C) Cluster similar data points together.
D) Reduce the dimensionality of image data.

A) Are based on tree-based models like Decision Trees and Random Forests.
B) Are based on distance calculations or gradient descent, such as SVM and Neural Networks.
C) Are used for association rule learning.
D) Are used for clustering only.

A) As the number of features grows, the data becomes increasingly sparse, making it harder to find meaningful patterns.
B) There are never enough features to train a good model.
C) Dimensionality reduction always improves model performance.
D) All datasets should have as many features as possible.

A) Make models more complex to fit the training data better.
B) Speed up the training time of a model.
C) Prevent overfitting by adding a penalty term to the loss function that discourages complex models.
D) Increase the variance of a model.

A) All features having non-zero weights.
B) Sparse models where the weights of less important features are driven to zero, effectively performing feature selection.
C) Increased model complexity.
D) A decrease in model interpretability.

A) The input features of the model.
B) Configuration settings for the learning algorithm that are not learned from the data and must be set prior to training (e.g., learning rate, K in K-NN).
C) The parameters that the model learns during training (e.g., weights in a neural network).
D) The output predictions of the model.

A) Cleaning the raw data.
B) Searching for the best combination of hyperparameters that results in the best model performance.
C) Training the model's internal weights.
D) Deploying the final model.

A) Randomly sampling hyperparameter combinations from a distribution.
B) Ignoring hyperparameters altogether.
C) Using a separate neural network to predict the best hyperparameters.
D) Exhaustively searching over a specified set of hyperparameter values.

A) Halting the training process when performance on a validation set starts to degrade, indicating the onset of overfitting.
B) Starting the training process later than scheduled.
C) Using a very small learning rate.
D) Stopping the training after a fixed, very short number of epochs.

A) Fully connected layers, where each neuron in one layer is connected to every neuron in the next layer.
B) Convolutional layers for processing images.
C) Recurrent layers for processing sequences.
D) A single layer of neurons.

A) Unsupervised learning problems.
B) Regression problems.
C) Binary classification problems.
D) Multi-class classification problems, as it outputs a probability distribution over the possible classes.

A) Does not require any hyperparameters.
B) Is guaranteed to find the global minimum for any function.
C) Is only used for unsupervised learning.
D) Combines the advantages of two other extensions of stochastic gradient descent, AdaGrad and RMSProp.

A) Improve the stability and speed of neural network training by normalizing the inputs to each layer.
B) Normalize the entire dataset before feeding it into the network.
C) Increase the batch size during training.
D) Replace the need for an activation function.

A) Classification model on a set of test data for which the true values are known.
B) Clustering algorithm's group assignments.
C) Dimensionality reduction technique's effectiveness.
D) Regression model's accuracy.

A) The model correctly predicted the negative class.
B) The model correctly predicted the positive class.
C) The model incorrectly predicted the positive class.
D) The model incorrectly predicted the negative class.

A) The learning rate is set too high.
B) The model is too complex for the data.
C) The features are not scaled properly.
D) One class in the training data has significantly more examples than another, which can bias the model.

A) Combines all classes into one.
B) Ignores the minority class completely.
C) Generates synthetic examples for the minority class to balance the dataset.
D) Deletes examples from the majority class at random.

A) It is only used for clustering unlabeled data.
B) It is a simpler and less powerful approach.
C) It requires a fully labeled dataset for training.
D) It learns by interacting with an environment and receiving rewards or penalties for actions, without a labeled dataset.

A) A policy that tells an agent what action to take under what circumstances by learning a value function.
B) A clustering of possible actions.
C) The principal components of a state space.
D) A decision tree for classification.

A) Reducing the dimensionality of word vectors.
B) Sentiment analysis, where text is classified as positive, negative, or neutral.
C) Generating new, original text without any input.
D) Grouping similar news articles without labels.

A) Are a type of clustering algorithm.
B) Represent words as simple one-hot encoded vectors.
C) Represent words as dense vectors in a continuous space, capturing semantic meaning.
D) Are used only for image classification.

A) Two identical Convolutional Neural Networks.
B) An Encoder and a Decoder for compression.
C) A Generator and a Discriminator, which are trained in opposition to each other.
D) A single, large Regression network.

A) Discriminating between real and fake data.
B) Classifying input images into categories.
C) Creating new, synthetic data that is indistinguishable from real data.
D) Reducing the dimensionality of the input.

A) Binary classifier that tries to correctly label data as real (from the dataset) or fake (from the generator).
B) Regression model predicting a continuous value.
C) Dimensionality reduction technique.
D) Clustering algorithm grouping similar images.