Font size
WorksheetsMachine Learning Concepts Worksheet
Total questions: 60
Worksheet time: 33mins
Which of the following is a supervised learning algorithm?
K-Means
Apriori
Decision Tree
PCA
What is the main goal of unsupervised learning?
Predict labels
Reduce dimensionality
Find patterns in data
Minimize loss
Which technique is used to prevent overfitting in machine learning?
A) Cross-validation
B) Gradient boosting
C) Early initialization
D) Grid sampling
Which algorithm is best suited for solving a regression problem?
Logistic Regression
Decision Tree Regressor
Naive Bayes
K-Means
Which of the following is not a distance metric?
Manhattan
Euclidean
Cosine
Entropy
Which concept explains the tradeoff between bias and variance?
Gradient descent
Regularization
Bias-Variance Tradeoff
Dimensionality reduction
Which of these is not a type of machine learning?
Reinforcement
Supervised
Heuristic
Unsupervised
Which ML technique would be most appropriate for detecting credit card fraud?
Regression
Classification
Which method is used for feature selection?
Dropout
Random Forest Feature Importance
Bagging
Pooling
Which activation function is used in binary classification?
ReLU
Tanh
Softmax
Sigmoid
Which library is mainly used for data manipulation in Python?
Matplotlib
Pandas
NumPy
Seaborn
Gradient Descent minimizes which of the following?
Accuracy
Precision
Cost Function
Recall
In random forest, how is randomness introduced?
Randomly selecting data samples and features
Random activation functions
Random weights
Random output threshold
What is the vanishing gradient problem?
Gradients explode to infinity
Gradients become too small to update weights
Weights vanish
Model accuracy drops suddenly.
Which of the following is not a hyperparameter in a decision tree?
Max depth
Min samples split
Learning rate
Criterion
Which ML technique is used in recommendation systems?
Classification
Clustering
Collaborative Filtering
Regression
The ROC curve is plotted with:
Precision vs Recall
TPR vs FPR
Accuracy vs Precision
F1 vs Recall
Which library in Python is used for deep learning?
NumPy
Matplotlib
TensorFlow
Pandas
Which evaluation metric measures the area under the ROC curve?
Precision
Recall
AUC
F1-score
What is the function of the loss function in training a model?
Improve model accuracy
Measure prediction error
Stop overfitting
Tune hyperparameters
Which is true about the Naive Bayes algorithm?
Assumes features are dependent
Requires normalization
Works poorly with categorical data
Assumes conditional independence
Which ML algorithm works on the principle of majority voting?
Logistic Regression
KNN
SVM
Naive Bayes
Which dimensionality reduction technique is linear?
t-SNE
PCA
UMAP
Autoencoder
A ride-sharing app wants to estimate the arrival time of a vehicle. Which type of ML problem is this?
Classification
Reinforcement Learning
Regression
Clustering
Which AI technique powers voice assistants like Siri or Alexa?
Clustering
Natural Language Processing (NLP)
Reinforcement Learning
Computer Vision
Which algorithm would be best for customer segmentation in marketing?
Decision Tree
K-Means
Linear Regression
Naive Bayes
In email spam detection, what kind of learning problem is being solved?
Clustering
Regression
Classification
Reinforcement Learning
An e-commerce website recommends products based on user behavior. What is this an example of?
Classification
Regression
Collaborative Filtering
Clustering
Which technique would be most suitable for detecting anomalies in network traffic?
Clustering
Regression
Classification
Supervised Learning
In facial recognition, what type of AI is being used?
NLP
Computer Vision
Reinforcement Learning
Genetic Algorithm
Which method would be used to forecast stock prices?
Classification
Regression
Clustering
Reinforcement Learning
A self-driving car deciding to stop or go is an example of:
Supervised Learning
Classification
Reinforcement Learning
Clustering
Which ML technique is typically used in optical character recognition (OCR)?
Clustering
CNN
RNN
SVM
Which of the following techniques can be used to handle vanishing gradients in deep neural networks?
Batch Normalization
Weight Initialization
ReLU Activation
Max Pooling
What is the main advantage of using residual connections in ResNets?
Reduces memory usage
Avoids overfitting
Enables training of deeper networks
Reduces number of parameters
In a Transformer model, which component is responsible for modeling sequence order?
Positional Encoding
Multi-head Attention
Feedforward Layer
Layer Normalization
Which reinforcement learning algorithm uses both a policy and a value function?
Q-Learning
SARSA
Deep Q Network
Actor-Critic
Which of the following is true about LSTM networks?
What is the key idea behind attention mechanisms in NLP models?
Compress sequences
Focus on important parts of the input
Encode sequence length
Reduce overfitting
Which algorithm is most suitable for solving multi-armed bandit problems?
Q-Learning
Softmax Classifier
Upper Confidence Bound (UCB)
Monte Carlo Tree Search
In Variational Autoencoders (VAEs), what is the role of the KL divergence term in the loss function?
Encourage sparsity
Ensure reconstruction accuracy
Regularize latent space distribution
Minimize overfitting
What differentiates Gradient Boosting from Bagging techniques like Random Forests?
Use of decision trees
Parallel training
Sequential model training using residuals
Sampling with replacement
Which property does the softmax function NOT satisfy?
Outputs are between 0 and 1
Outputs sum to 1
Preserves order of inputs
Used in multi-class classification
What is the role of the “critic” in Actor-Critic RL methods?
Suggests next action
Estimates value function
Optimizes the environment
Calculates policy gradient
Which concept is central to the success of self-supervised learning?
Data Augmentation
Ensemble Learning
Which of the following methods is used to interpret predictions from black-box models?
Logistic Regression
LIME
Grid Search
Q-Learning
Which of the following techniques can be used to handle vanishing gradients in deep neural networks?
Batch Normalization
Weight Initialization
ReLU Activation
Max Pooling
What is the main advantage of using residual connections in ResNets?
Reduces memory usage
Avoids overfitting
Enables training of deeper networks
Reduces number of parameters
In a Transformer model, which component is responsible for modeling sequence order?
Positional Encoding
Multi-head Attention
Feedforward Layer
Layer Normalization
Which reinforcement learning algorithm uses both a policy and a value function?
Q-Learning
SARSA
Deep Q Network
Actor-Critic
Which of the following is true about LSTM networks?
They can only handle fixed-length sequences
They suffer from vanishing gradients more than RNNs
They use gating mechanisms to control memory flow
They require convolutional layers to function
What is the key idea behind attention mechanisms in NLP models?
Compress sequences
Focus on important parts of the input
Encode sequence length
Which algorithm is most suitable for solving multi-armed bandit problems?
Q-Learning
Softmax Classifier
Upper Confidence Bound (UCB)
Monte Carlo Tree Search
In Variational Autoencoders (VAEs), what is the role of the KL divergence term in the loss function?
Encourage sparsity
Ensure reconstruction accuracy
Regularize latent space distribution
Minimize overfitting
What differentiates Gradient Boosting from Bagging techniques like Random Forests?
Use of decision trees
Parallel training
Sequential model training using residuals
Sampling with replacement
Which property does the softmax function NOT satisfy?
Outputs are between 0 and 1
Outputs sum to 1
Preserves order of inputs
Used in multi-class classification
What is the role of the “critic” in Actor-Critic RL methods?
Suggests next action
Estimates value function
Optimizes the environment
Calculates policy gradient
Which concept is central to the success of self-supervised learning?
Data Augmentation
Ensemble Learning
Overparameterization
One-hot Encoding
Which of the following methods is used to interpret predictions from black-box models?
Logistic Regression
LIME
Which AI approach focuses on mimicking the human brain structure and function?
Symbolic AI
Sub-symbolic AI
Connectionism
Cybernetics
