Deep Learning Unit-I

Which of the following is the main goal of a Support Vector Machine (SVM)?

• Minimize error

• Maximize margin

• Minimize variance

• Minimize computational cost

Which of the following is true about a Perceptron?

• It can only solve linearly separable problems

• It is equivalent to a Support Vector Machine

• It works well with multi-class problems without modification

• It does not use an activation function

In logistic regression, what does the logistic function (sigmoid) do?

• Maps values to a range between 0 and 1

• Transforms input into linear combinations

• Reduces bias in the model

• Maximizes the margin between classes

Which method is used in SVM to find the optimal hyperplane?

• Gradient Descent

• Stochastic Gradient Descent

• Quadratic Programming

• Newton's Method

What is the decision boundary in a linear model like Logistic Regression?

• A hyperplane

• A curve

• A random line

• A cluster of points

Which of the following loss functions is used for logistic regression?

• Mean Squared Error

• Cross-Entropy Loss

• Hinge Loss

• L2 Loss

What does the kernel trick in SVMs allow us to do?

• Increase the model's complexity

• Map the input features into higher-dimensional space

• Reduce the training time

• Normalize the input features

What is the activation function used in the Perceptron model?

• ReLU

• Sigmoid

• Step function

• Tanh

Which of the following is an advantage of using SVMs?

• Can handle non-linear decision boundaries

• Requires fewer training examples than decision trees

• No need for feature scaling

• Works best with a large number of classes

What does regularization in logistic regression help prevent?

• Overfitting

• Underfitting

• Redundant features

• Outliers in the dataset