Q. In the context of CNNs, what does 'stride' refer to?
A.
The number of filters used
B.
The step size of the filter during convolution
C.
The depth of the network
D.
The size of the input image
Show solution
Solution
Stride refers to the step size of the filter during convolution, determining how much the filter moves across the input image.
Correct Answer:
B
— The step size of the filter during convolution
Learn More →
Q. In which application are CNNs most commonly used?
A.
Natural Language Processing
B.
Image Recognition
C.
Time Series Forecasting
D.
Reinforcement Learning
Show solution
Solution
CNNs are most commonly used in image recognition tasks due to their ability to capture spatial hierarchies in images.
Correct Answer:
B
— Image Recognition
Learn More →
Q. In which scenario would you typically use a CNN?
A.
Predicting stock prices
B.
Classifying images
C.
Analyzing text data
D.
Clustering customer segments
Show solution
Solution
CNNs are primarily used for classifying images due to their ability to capture spatial hierarchies in visual data.
Correct Answer:
B
— Classifying images
Learn More →
Q. What does CNN stand for in the context of deep learning?
A.
Convolutional Neural Network
B.
Cyclic Neural Network
C.
Complex Neural Network
D.
Conditional Neural Network
Show solution
Solution
CNN stands for Convolutional Neural Network, which is a class of deep neural networks commonly used for analyzing visual imagery.
Correct Answer:
A
— Convolutional Neural Network
Learn More →
Q. What is overfitting in the context of deep learning?
A.
When the model performs well on training data but poorly on unseen data
B.
When the model performs equally on training and test data
C.
When the model is too simple to capture the underlying patterns
D.
When the model has too many parameters
Show solution
Solution
Overfitting occurs when a model learns the training data too well, including noise, resulting in poor performance on unseen data.
Correct Answer:
A
— When the model performs well on training data but poorly on unseen data
Learn More →
Q. What is overfitting in the context of training CNNs?
A.
When the model performs well on training data but poorly on unseen data
B.
When the model is too simple to capture the underlying patterns
C.
When the model has too few parameters
D.
When the model is trained on too much data
Show solution
Solution
Overfitting occurs when the model performs well on training data but poorly on unseen data, indicating it has learned noise rather than the underlying pattern.
Correct Answer:
A
— When the model performs well on training data but poorly on unseen data
Learn More →
Q. What is the main advantage of using CNNs over traditional machine learning methods for image classification?
A.
They require less data
B.
They automatically learn features from data
C.
They are easier to implement
D.
They are faster to train
Show solution
Solution
CNNs automatically learn features from data, which eliminates the need for manual feature extraction that is often required in traditional methods.
Correct Answer:
B
— They automatically learn features from data
Learn More →
Q. What is the main purpose of the softmax function in a CNN?
A.
To normalize the output to a probability distribution
B.
To reduce dimensionality
C.
To apply a non-linear transformation
D.
To perform convolution
Show solution
Solution
The softmax function normalizes the output of the final layer to a probability distribution over multiple classes in classification tasks.
Correct Answer:
A
— To normalize the output to a probability distribution
Learn More →
Q. What is the primary advantage of using transfer learning in CNNs?
A.
It requires less data to train
B.
It speeds up the training process
C.
It improves model accuracy
D.
All of the above
Show solution
Solution
Transfer learning allows models to leverage pre-trained weights, which can lead to faster training, improved accuracy, and reduced data requirements.
Correct Answer:
D
— All of the above
Learn More →
Q. What is the purpose of the pooling layer in a CNN?
A.
To increase the dimensionality of the data
B.
To reduce the spatial size of the representation
C.
To apply non-linear transformations
D.
To connect neurons in the network
Show solution
Solution
The pooling layer reduces the spatial size of the representation, which decreases the number of parameters and computation in the network.
Correct Answer:
B
— To reduce the spatial size of the representation
Learn More →
Q. What is the role of dropout in a CNN?
A.
To increase the number of neurons
B.
To prevent overfitting
C.
To enhance feature extraction
D.
To speed up training
Show solution
Solution
Dropout is a regularization technique used to prevent overfitting by randomly setting a fraction of input units to zero during training.
Correct Answer:
B
— To prevent overfitting
Learn More →
Q. What is the role of the fully connected layer in a CNN?
A.
To perform convolution operations
B.
To reduce dimensionality
C.
To connect every neuron in one layer to every neuron in the next layer
D.
To apply pooling
Show solution
Solution
The fully connected layer connects every neuron in one layer to every neuron in the next layer, allowing for integration of features learned by previous layers.
Correct Answer:
C
— To connect every neuron in one layer to every neuron in the next layer
Learn More →
Q. What is transfer learning in deep learning?
A.
Training a model from scratch on a new dataset
B.
Using a pre-trained model on a new but related task
C.
Fine-tuning a model on the same dataset
D.
Applying unsupervised learning techniques
Show solution
Solution
Transfer learning involves using a pre-trained model on a new but related task, which can save time and resources.
Correct Answer:
B
— Using a pre-trained model on a new but related task
Learn More →
Q. What is transfer learning in the context of CNNs?
A.
Training a model from scratch on a new dataset
B.
Using a pre-trained model on a new but related task
C.
Combining multiple models to improve performance
D.
Fine-tuning hyperparameters of a model
Show solution
Solution
Transfer learning involves using a pre-trained model on a new but related task, allowing for faster training and improved performance.
Correct Answer:
B
— Using a pre-trained model on a new but related task
Learn More →
Q. Which activation function is commonly used in CNNs?
A.
Sigmoid
B.
Tanh
C.
ReLU
D.
Softmax
Show solution
Solution
ReLU (Rectified Linear Unit) is commonly used in CNNs due to its ability to introduce non-linearity while being computationally efficient.
Correct Answer:
C
— ReLU
Learn More →
Q. Which evaluation metric is commonly used for image classification tasks?
A.
Mean Squared Error
B.
Accuracy
C.
F1 Score
D.
Confusion Matrix
Show solution
Solution
Accuracy is a commonly used evaluation metric for image classification tasks, measuring the proportion of correctly classified instances.
Correct Answer:
B
— Accuracy
Learn More →
Q. Which layer in a CNN is primarily responsible for feature extraction?
A.
Pooling layer
B.
Fully connected layer
C.
Convolutional layer
D.
Activation layer
Show solution
Solution
The Convolutional layer is responsible for feature extraction by applying filters to the input data.
Correct Answer:
C
— Convolutional layer
Learn More →
Q. Which of the following is a common activation function used in CNNs?
A.
Sigmoid
B.
ReLU
C.
Tanh
D.
Softmax
Show solution
Solution
ReLU (Rectified Linear Unit) is a common activation function used in CNNs due to its simplicity and effectiveness in training deep networks.
Correct Answer:
B
— ReLU
Learn More →
Q. Which of the following is a common evaluation metric for image classification tasks?
A.
Mean Squared Error
B.
Accuracy
C.
F1 Score
D.
Confusion Matrix
Show solution
Solution
Accuracy is a common evaluation metric for image classification tasks, measuring the proportion of correctly classified instances.
Correct Answer:
B
— Accuracy
Learn More →
Q. Which of the following is a common technique to prevent overfitting in CNNs?
A.
Increasing the learning rate
B.
Using dropout layers
C.
Reducing the number of layers
D.
Using a smaller batch size
Show solution
Solution
Using dropout layers is a common technique to prevent overfitting by randomly setting a fraction of input units to 0 during training.
Correct Answer:
B
— Using dropout layers
Learn More →
Showing 1 to 20 of 20 (1 Pages)
