COELE1-2

1. "Mini-batch Gradient Descent" is often preferred because it:

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

2. "Transfer Learning" in deep learning involves:

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

3. An "Autoencoder" is a type of neural network primarily used for:

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

4. The architecture of a typical autoencoder consists of:

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

5. In the context of model evaluation for classification, "Accuracy" is defined as:

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

6. "Precision" is an important metric when:

A) You are evaluating a regression model.
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 need a single metric that combines precision and recall.

7. "Recall" is an important metric when:

A) You are evaluating a clustering model.
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).
D) You need a single metric that combines precision and recall.

8. The "F1 Score" is:

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

9. For a regression model, the "Mean Squared Error" (MSE) measures:

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

10. The "ROC Curve" is a tool used to evaluate:

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

11. "Area Under the ROC Curve" (AUC) provides an aggregate measure of performance across all possible classification thresholds. A perfect model has an AUC of:

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

12. "K-fold Cross-Validation" is a technique used to:

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

13. In the K-Nearest Neighbors (K-NN) algorithm for classification, the class of a new data point is determined by:

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

14. The parameter 'K' in the K-NN algorithm:

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

15. "Principal Component Analysis" (PCA) works by:

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.

16. The first principal component in PCA is the direction in the feature space that:

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

17. "K-Means Clustering" aims to partition data into K clusters such that:

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

18. The "Elbow Method" is a heuristic used in K-Means to:

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

19. "Naive Bayes" classifiers are called "naive" because they:

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

20. "Logistic Regression" is fundamentally a:

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

21. The output of a logistic regression model is a value between 0 and 1, which represents the:

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

22. A "Random Forest" is an ensemble method that combines multiple:

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

23. The "bagging" technique in a Random Forest helps to:

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

24. "Gradient Boosting" machines (e.g., XGBoost) are ensemble methods that:

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

25. The term "feature engineering" refers to:

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

26. "One-hot encoding" is a preprocessing technique used to:

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

27. "Feature scaling" (e.g., normalization or standardization) is often crucial for algorithms that:

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

28. The "curse of dimensionality" refers to the problem that:

A) Dimensionality reduction always improves model performance.
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) All datasets should have as many features as possible.

29. "Regularization" is a technique used to:

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

30. L1 Regularization (Lasso) can often lead to:

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

31. "Hyperparameters" are:

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

32. The process of "Hyperparameter Tuning" involves:

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

33. "Grid Search" is a common method for hyperparameter tuning that involves:

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

34. "Early Stopping" is a form of regularization that works by:

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.

35. A "Vanilla" neural network, also known as a Multilayer Perceptron (MLP), is typically composed of:

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

36. The "softmax" activation function is commonly used in the output layer of a neural network for:

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

37. The "Adam" optimizer is an adaptive learning rate algorithm that is often preferred because it:

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

38. "Batch Normalization" is a technique used to:

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

39. The "confusion matrix" is a table that is used to describe the performance of a:

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

40. In a confusion matrix, the "true positives" are the cases where:

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

41. The problem of "imbalanced classes" occurs when:

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

42. A technique to address imbalanced classes is "SMOTE," which:

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

43. "Reinforcement Learning" differs from supervised and unsupervised learning in that:

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

44. "Q-Learning" is a popular algorithm in reinforcement learning that learns:

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

45. "Natural Language Processing" (NLP) often uses supervised learning for tasks like:

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

46. "Word Embeddings" (like Word2Vec) are techniques that:

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

47. A "Generative Adversarial Network" (GAN) consists of two networks:

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

48. The "Generator" in a GAN is responsible for:

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

49. The "Discriminator" in a GAN is essentially a:

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.

Created with That Quiz — a math test site for students of all grade levels.