MACHINE LEARNING A BAYESIAN AND
OPTIMIZATION PERSPECTIVE FINAL
PAPER FULL QUESTIONS AND CORRECT
ANSWERS PREMIUM STUDY SHEET
●● What is the goal of classification in machine learning?
Answer: To predict labels (classes) y given inputs x.
●● What does regression in machine learning focus on?
Answer: Learning predictors with real-valued outputs.
●● What is reinforcement learning?
Answer: A type of learning that involves a sequence of rewards with no
correct answers.
●● Define unsupervised learning.
Answer: A learning approach that aims to make sense of the data without
labeled outputs.
●● What is feature extraction?
Answer: The process of identifying attributes of the input that help in
classification.
, ●● What is the workflow of the classification process?
Answer: Input is provided, features are extracted, a machine learning
algorithm is run, and a predicted label is output.
●● What is the big idea behind machine learning algorithms?
Answer: They learn patterns between features and labels from data.
●● What is an example of a classification task?
Answer: Medical diagnosis, where symptoms are the input and diseases
are the classes.
●● What does Naïve Bayes assume about features?
Answer: It assumes that all features are independent effects of the label.
●● What is the purpose of training in machine learning?
Answer: To learn patterns from labeled data and test labels for unlabeled
data.
●● In the context of a spam filter, what is the input?
Answer: An email.
●● What does the output of a spam filter represent?
OPTIMIZATION PERSPECTIVE FINAL
PAPER FULL QUESTIONS AND CORRECT
ANSWERS PREMIUM STUDY SHEET
●● What is the goal of classification in machine learning?
Answer: To predict labels (classes) y given inputs x.
●● What does regression in machine learning focus on?
Answer: Learning predictors with real-valued outputs.
●● What is reinforcement learning?
Answer: A type of learning that involves a sequence of rewards with no
correct answers.
●● Define unsupervised learning.
Answer: A learning approach that aims to make sense of the data without
labeled outputs.
●● What is feature extraction?
Answer: The process of identifying attributes of the input that help in
classification.
, ●● What is the workflow of the classification process?
Answer: Input is provided, features are extracted, a machine learning
algorithm is run, and a predicted label is output.
●● What is the big idea behind machine learning algorithms?
Answer: They learn patterns between features and labels from data.
●● What is an example of a classification task?
Answer: Medical diagnosis, where symptoms are the input and diseases
are the classes.
●● What does Naïve Bayes assume about features?
Answer: It assumes that all features are independent effects of the label.
●● What is the purpose of training in machine learning?
Answer: To learn patterns from labeled data and test labels for unlabeled
data.
●● In the context of a spam filter, what is the input?
Answer: An email.
●● What does the output of a spam filter represent?
