COELE1-2 - Quiz

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

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.

2. "Transfer Learning" in deep learning involves:

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.

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

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

4. The architecture of a typical autoencoder consists of:

A) A convolutional layer followed by an RNN layer.
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) Only a single layer of perceptrons.

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

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

6. "Precision" is an important metric when:

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

7. "Recall" is an important metric when:

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) You are evaluating a clustering model.
D) The cost of false positives is high (e.g., in spam detection).

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 variance of the input features.
B) The accuracy of a classification model.
C) The total number of misclassified instances.
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 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.

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

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

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

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

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

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

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

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

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

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

A) Captures the greatest possible variance in the data.
B) Is perpendicular to all other components.
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 between-cluster variance is minimized.
B) The data is perfectly classified into known labels.
C) The data is projected onto a single dimension.
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) Initialize the cluster centroids.
C) Determine the learning rate for gradient descent.
D) Help choose the optimal number of clusters K by looking for a "bend" in the plot of within-cluster variance.

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

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

20. "Logistic Regression" is fundamentally a:

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

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

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

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

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

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

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

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

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

25. The term "feature engineering" refers to:

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

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

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

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

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

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

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) Dimensionality reduction always improves model performance.
D) There are never enough features to train a good model.

29. "Regularization" is a technique used to:

A) Increase the variance of a model.
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) Make models more complex to fit the training data better.

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

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.

31. "Hyperparameters" are:

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

32. The process of "Hyperparameter Tuning" involves:

A) Training the model's internal weights.
B) Cleaning the raw data.
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) Randomly sampling hyperparameter combinations from a distribution.
B) Exhaustively searching over a specified set of hyperparameter values.
C) Ignoring hyperparameters altogether.
D) Using a separate neural network to predict the best hyperparameters.

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

A) Using a very small learning rate.
B) Starting the training process later than scheduled.
C) Halting the training process when performance on a validation set starts to degrade, indicating the onset of overfitting.
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) Recurrent layers for processing sequences.
C) Fully connected layers, where each neuron in one layer is connected to every neuron in the next layer.
D) Convolutional layers for processing images.

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

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

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

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.

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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.

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

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.

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

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

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