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

A) Training a model from scratch on every new problem.
B) 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.
C) Forgetting everything a model has learned.
D) Using only unsupervised learning techniques.

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

A) The output predictions of the model.
B) The input features of the model.
C) 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).
D) The parameters that the model learns during training (e.g., weights in a neural network).

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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