MCQs

Disclaimer:

Educational Use Only:

The website links provided are to be used strictly for educational and knowledge-gaining purposes only.

No Misuse:

Do not misuse the links or the information contained therein for any commercial, illegal, or unethical activities.

Respect for Websites and Businesses:

Respect the intellectual property and business interests of the websites and their owners. Do not attempt to alter, copy, or redistribute content without proper authorization.

Acknowledge Value of Knowledge:

Recognize and value the knowledge and resources shared by these websites. Use the information responsibly and ethically.

No Endorsement:

The provision of these links does not imply endorsement of the linked sites, their content, or their products and services. Users are responsible for evaluating the suitability and accuracy of the information.

User Responsibility:

By accessing these links, you agree to use the information at your own risk and to comply with all applicable laws and website terms of use.

Overall MCQs for Reference

What is the main objective of model optimization in artificial intelligence?
a. Maximizing color contrast
b. Increasing computational cost
c. Improving model performance
d. Reducing training data
Answer: c. Improving model performancekeymakr+1
Which technique is commonly used to minimize the loss function during model training?
a. Simulated annealing
b. Gradient descent
c. Random forest
d. K-means clustering
Answer: b. Gradient descentgeeksforgeeks+1
What is the purpose of regularization in machine learning models?
a. Increase model complexity
b. Prevent overfitting
c. Reduce dataset size
d. Add more layers
Answer: b. Prevent overfittingsimplilearn+1
Which of these is NOT a hyperparameter optimization technique?
a. Grid search
b. Random search
c. Bayesian optimization
d. Data augmentation
Answer: d. Data augmentationkeymakr+1
What does L1 regularization penalize in model weights?
a. Large negative weights only
b. Large positive weights only
c. Sum of the absolute values of weights
d. Activation function
Answer: c. Sum of the absolute values of weightsgeeksforgeeks+1
Which optimization algorithm is stochastic and often used for large datasets?
a. Simplex algorithm
b. Stochastic gradient descent (SGD)
c. A* search
d. Minimax
Answer: b. Stochastic gradient descent (SGD)keymakr+1
Which parameter is tuned during hyperparameter optimization?
a. Training data labels
b. Activation function type
c. Learning rate
d. Input features
Answer: c. Learning rategeeksforgeeks+1
What is early stopping in model training?
a. Stopping when GPU overheats
b. Stopping training when validation performance degrades
c. Reducing the batch size
d. Increasing epochs to infinity
Answer: b. Stopping training when validation performance degradessimplilearn+1
What does model pruning involve?
a. Adding more nodes
b. Removing unnecessary parameters
c. Increasing dataset size
d. Repeating training
Answer: b. Removing unnecessary parametersnetguru+1
Which method reduces model size by lowering parameter precision?
a. Hashing
b. Quantization
c. Clustering
d. Pooling
Answer: b. Quantizationnetguru
Which term measures how well a model generalizes to new data?
a. Training accuracy
b. Overfitting score
c. Model generalization
d. Hyperparameter score
Answer: c. Model generalizationsimplilearn+1
What is the main advantage of using dropout in neural networks?
a. Reduces computation time
b. Increases model size
c. Helps prevent overfitting
d. Ignores input features
Answer: c. Helps prevent overfittingkeymakr+1
Which optimization method uses neighborhood search to escape local optima?
a. Hill climbing
b. Random search
c. Simulated annealing
d. Genetic algorithms
Answer: c. Simulated annealinggeeksforgeeks+1
What does “batch size” control in neural network training?
a. Number of hidden layers
b. Learning rate adjustment method
c. Number of samples processed before update
d. Dimensionality of input
Answer: c. Number of samples processed before updategeeksforgeeks
Which technique is best for reducing input features while retaining information?
a. Data augmentation
b. Principal Component Analysis (PCA)
c. Cross-validation
d. Bagging
Answer: b. Principal Component Analysis (PCA)geeksforgeeks
Grid search hyperparameter tuning is:
a. Random sampling of parameters
b. Exhaustively searching parameter combinations
c. Increasing dataset size
d. Decreasing number of epochs
Answer: b. Exhaustively searching parameter combinationskeymakr+1
What is an epoch in machine learning training?
a. One backward pass only
b. A set of hyperparameter values
c. One complete pass through the training data
d. Size of the dataset
Answer: c. One complete pass through the training datageeksforgeeks
Which umbrella term refers to restricting model complexity to improve generalization?
a. Overfitting
b. Regularization
c. Data wrangling
d. Feature engineering
Answer: b. Regularizationsimplilearn+1
What type of optimization problem is solved by linear programming?
a. Nonlinear objective
b. Linear objective
c. Classification
d. Regression only
Answer: b. Linear objectivescribd+1
Which optimization method mimics evolution in its search process?
a. Hill climbing
b. Simulated annealing
c. Genetic algorithm
d. Random forest
Answer: c. Genetic algorithmgeeksforgeeks
What is “model quantization” used for?
a. Increasing model compute
b. Reducing parameter bit-width for efficiency
c. Enhancing dataset size
d. Selecting activation functions
Answer: b. Reducing parameter bit-width for efficiencynetguru
Why is cross-validation used in model optimization?
a. To increase bias
b. To assess how the model generalizes to unseen data
c. To reduce input size
d. To increase output size
Answer: b. To assess how the model generalizes to unseen datasimplilearn+1
What does “learning rate decay” help prevent?
a. Gradient vanishing
b. Model pruning
c. Overfitting
d. Overshooting minima
Answer: d. Overshooting minimageeksforgeeks
Which technique is NOT for model optimization?
a. Dropout
b. L1 regularization
c. Transfer learning
d. Data leakage
Answer: d. Data leakagekeymakr
What is objective function in optimization?
a. Function to minimize or maximize
b. Regularizer term
c. Dataset label function
d. Noise injection function
Answer: a. Function to minimize or maximizescribd+1
Which of these is a regularization technique for neural networks?
a. Max pooling
b. Dropout
c. PCA
d. AdaBoost
Answer: b. Dropoutsimplilearn+1
Which metric is most relevant for imbalanced datasets?
a. Accuracy
b. Recall
c. Training loss
d. Random error
Answer: b. Recallkeymakr
What is the role of “momentum” in optimization algorithms?
a. Increase model complexity
b. Accelerate convergence and avoid local minima
c. Reduce dataset features
d. Convert loss to accuracy
Answer: b. Accelerate convergence and avoid local minimageeksforgeeks
Which algorithm updates weights using loss derivatives?
a. Gradient descent
b. Naive Bayes
c. K-Nearest Neighbors
d. Apriori
Answer: a. Gradient descentlearncbse+1
What is “hyperparameter tuning”?
a. Encrypting training data
b. Optimizing model performance through parameter adjustment
c. Reducing data size
d. Feature expansion
Answer: b. Optimizing model performance through parameter adjustmentigmguru+1
What is “overfitting”?
a. The model fits noise in training data
b. Underutilizing memory resources
c. High training error
d. Increased validation accuracy only
Answer: a. The model fits noise in training datasimplilearn+1
Which function is used to measure prediction errors in regression models?
a. Activation function
b. Loss function
c. Softmax
d. Decision tree
Answer: b. Loss functiongeeksforgeeks
Which method reduces model variance?
a. Increasing number of parameters
b. Bagging
c. Increasing epochs
d. Clustering
Answer: b. Baggingkeymakr+1
Which architecture is commonly optimized through backpropagation?
a. Decision tree
b. Neural network
c. Clustering tree
d. Association rule mining
Answer: b. Neural networktestbook
What is the purpose of Bayesian optimization?
a. Grid search parameter tuning
b. Probabilistic modeling of parameter space for efficient optimization
c. Adding random noise
d. Reducing dimensionality
Answer: b. Probabilistic modeling of parameter space for efficient optimizationkeymakr+1
Which technique adjusts model complexity using a penalty factor?
a. Data shuffling
b. Regularization
c. Boosting
d. Bagging
Answer: b. Regularizationsimplilearn+1
What is “validation loss” used for in optimization?
a. Optimizing training speed
b. Monitoring generalization to prevent overfitting
c. Tracking memory usage
d. Counting epochs
Answer: b. Monitoring generalization to prevent overfittingsimplilearn+1
Which is a method for reducing overfitting?
a. Increasing model parameters
b. Early stopping
c. Zero regularization
d. Infinite batch size
Answer: b. Early stoppingkeymakr+1
What does “model tuning” directly affect?
a. Model hyperparameters
b. Data collection method
c. Only model visualization
d. Loss function definition
Answer: a. Model hyperparameterskeymakr+1
What is the risk of overly complex models?
a. Underfitting
b. Overfitting
c. Model quantization
d. Reduced regularization
Answer: b. Overfittingsimplilearn+1
Which optimizer is based on adaptive learning rates?
a. SGD
b. Adam
c. KNN
d. Random forest
Answer: b. Adamgeeksforgeeks
What is the role of “weight decay”?
a. Prevent increasing weights
b. Increase overfitting
c. Reduce feature space
d. Increase learning rate
Answer: a. Prevent increasing weightskeymakr+1
Which method splits the dataset into multiple subsets for model assessment?
a. Cross-validation
b. Bagging
c. Dropout
d. Pooling
Answer: a. Cross-validationsimplilearn+1
What is the main benefit of “transfer learning” in model optimization?
a. Forces retraining on large data
b. Uses pre-trained features for faster optimization
c. Increases model size
d. Adds more regularization
Answer: b. Uses pre-trained features for faster optimizationgeeksforgeeks
Which type of machine learning problem benefits most from the F1-score metric?
a. Balanced datasets
b. Imbalanced classification
c. Regression
d. Clustering
Answer: b. Imbalanced classificationkeymakr
What is “feature selection” in model optimization?
a. Choosing important input variables
b. Selecting target labels
c. Tuning hyperparameters
d. Reducing learning rate
Answer: a. Choosing important input variablesgeeksforgeeks+1
Why use the mini-batch gradient descent?
a. Only for linear regression
b. Balance between computation and convergence speed
c. Always slower
d. Never used for optimization
Answer: b. Balance between computation and convergence speedgeeksforgeeks
Which term describes the difference between training and validation error?
a. Dataset imbalance
b. Generalization gap
c. Gradient flow
d. Loss function choice
Answer: b. Generalization gapsimplilearn+1
What helps combat vanishing gradients in deep neural networks?
a. Dropout
b. ReLU activation function
c. Increasing batch size
d. Max pooling
Answer: b. ReLU activation functiongeeksforgeeks
What is “ensemble learning” in AI model optimization?
a. Using a single model always
b. Combining multiple models for better performance
c. Pruning neural networks
d. Regularization technique
Answer: b. Combining multiple models for better performancekeymakr+1rformance**geeksforgeeks+1

Blog

Systemethic Technologies

Leave a Reply Cancel reply

Useful links

Other Terms & Policies

Contact Info