FMSF86/FMSF90 Statistical learning

What does the term "supervised learning" refer to in machine learning?

• Learning with a teacher or supervisor

• Learning with labeled data

• Learning in a classroom setting

• Learning with minimal guidance

What are some common algorithms used in unsupervised learning?

• Naive Bayes, K-nearest neighbors, AdaBoost

• Linear regression, Decision tree, Support vector machine

• K-means, hierarchical clustering, DBSCAN

• Principal component analysis, Logistic regression, Random forest

Explain the concept of hyperparameter tuning in machine learning.

• It involves adjusting the model complexity during model training

• It is the process of fine-tuning model parameters after training

• It focuses on reducing the number of features in a model

• It refers to the selection of the most important features in a dataset

Explain the term "bias" in the context of machine learning models.

• Bias is the error introduced by a model due to overfitting

• Bias is the difference between predicted and actual values in the training data

• Bias represents the tendency of a model to systematically underpredict or overpredict

• Bias is a measure of the model's sensitivity to changes in input features

How is linear regression used in machine learning?

• Linear regression is used to establish a non-linear relationship between input and output variables

• Linear regression is used to classify data into categories

• Linear regression is used to predict discrete values

• Linear regression is used to predict continuous values

How does adding more variables to a linear regression model affect the R-squared value?

• It always increases

• It always decreases

• It remains the same

• It becomes zero

What is logistic regression?

• Logistic regression is just a synonym for linear regression used for classification

• Logistic regression is linear regression but with LASSO penalty on sum(log(beta))

• Logistic regression is a regression model for classification that avoids impossible probability estimates by a sigmoid transformation

• Logistic regression is a black-box regression model for classification that avoids impossible probability estimates by penalizing prob<0 or prob>1.

Explain the concept of decision trees and how they are used in machine learning.

• Decision trees are used in machine learning to create linear regression models

• Decision trees are used in machine learning to make predictions or decisions by creating a tree-like model of decisions based on features of the data.

• Decision trees are used in machine learning to calculate statistical probabilities

• Decision trees are used in machine learning to represent the chronological order of events

What is the concept of bagging, and how does it improve model performance

• Bagging involves training a single model on multiple subsets of the training data, reducing overfitting

• Bagging is a technique to increase model complexity by adding more layers to the architecture

• Bagging combines predictions from multiple models to reduce variance and improve generalization

• Bagging is used to decrease the number of features in a machine learning model

What is the difference between bagging and boosting in ensemble learning?

• Bagging involves training multiple models sequentially, while boosting trains them in parallel

• Bagging aims to reduce model variance, while boosting focuses on reducing bias

• Bagging combines predictions from multiple models, while boosting combines predictions by assigning weights to models

• Bagging is only applicable to decision tree models, while boosting can be applied to any algorithm