Machine Learning Exam
What are the two categories of machine learning? correct answers Supervised
Learning and Unsupervised Learning
What is Supervised Learning? correct answers -Goal is to generalize from training
examples
-predict "labels" given "features"
-regression, binary classification, multiclass classfiication, ranking
What is unsupervised learning? correct answers -no labeled training examples.. so just
try to understand the distribution of your data
-density estimation, clustering, classfication
What is are the three label spaces? correct answers binary classfication, multi-class
classification, regression
What is the No Free Lunch Theorem? correct answers No single machine learning
algorithm is always the best
What is the purpose of a loss function? correct answers To find a function h within the
hypothesis class that makes the fewest mistakes within our training data
What is zero-one loss function? correct answers It counts how many mistakes a
hypothesis function h makes on the training set
What is squared loss function? correct answers -typically used in regression settings
-the loss suffered is always non-negative
-the loss suffered grows quadratically with the absolute mispredicted amount
What is absolute loss function? correct answers -the suffered loss grows linearly with
the mispredictions it is more suitable for noisy data
What is overfitting? correct answers Overfitting, or *high variance*, is caused by a
hypothesis function that fits the available data but does not generalize well to predict
new data.
What is the purpose of the validation set in machine learning? correct answers It is
used to find the and optimize the best model to solve a given problem so tune the
hyperparameters.
What is the K-NN algorithm? correct answers It is k-nearest neighbors algorithm that
looks at a certain k inputs around a seed point then based on the proximity (distance) to
What are the two categories of machine learning? correct answers Supervised
Learning and Unsupervised Learning
What is Supervised Learning? correct answers -Goal is to generalize from training
examples
-predict "labels" given "features"
-regression, binary classification, multiclass classfiication, ranking
What is unsupervised learning? correct answers -no labeled training examples.. so just
try to understand the distribution of your data
-density estimation, clustering, classfication
What is are the three label spaces? correct answers binary classfication, multi-class
classification, regression
What is the No Free Lunch Theorem? correct answers No single machine learning
algorithm is always the best
What is the purpose of a loss function? correct answers To find a function h within the
hypothesis class that makes the fewest mistakes within our training data
What is zero-one loss function? correct answers It counts how many mistakes a
hypothesis function h makes on the training set
What is squared loss function? correct answers -typically used in regression settings
-the loss suffered is always non-negative
-the loss suffered grows quadratically with the absolute mispredicted amount
What is absolute loss function? correct answers -the suffered loss grows linearly with
the mispredictions it is more suitable for noisy data
What is overfitting? correct answers Overfitting, or *high variance*, is caused by a
hypothesis function that fits the available data but does not generalize well to predict
new data.
What is the purpose of the validation set in machine learning? correct answers It is
used to find the and optimize the best model to solve a given problem so tune the
hyperparameters.
What is the K-NN algorithm? correct answers It is k-nearest neighbors algorithm that
looks at a certain k inputs around a seed point then based on the proximity (distance) to
