Raccolta di possibili domande di teoria

Page 1

7/2/2017 1.Describe and compare ridge regression and lasso. 2.Define the VC dimension and describe the importance and usefulness of VC dimension in machine learning. 3.Describe the policy iteration algorithm. 7/7/2016 1.Describe the supervised learning technique denominated Support Vector Machines for classification problems. 2.Describe the supervised learning technique denominated logistic regression for classification problems. 3.Describe the differences existing between the Montecarlo and the Temporal Difference methods in the model–free estimation of a value function for a given policy. 9/9/2016 1.Describe the supervised learning technique denominated K–Nearest Neighbor for classification problems. 2.Describe the Bias-Variance tradeoff for regression problems. Explain how is it possible to evaluate the bias-variance tradeoff by looking at the train error and at the test error. 3.Describe the difference between on-policy and off-policy reinforcement learning techniques. Make an example of an on-policy algorithm and an example of an off-policy algorithm. 19/7/2016 1.Describe the supervised learning technique denominated Ridge regression for regression problems. 2.Describe the unsupervised learning technique denominated K–means for clustering problems. 3.Describe the policy iteration technique for control problems on Markov Decision Processes. 26/9/2016 1.Describe the supervised learning technique denominated Support Vector Machines (SVMs) for classification problems. 2.Describe the UCB1 algorithm for multi-armed bandit problems. 3.Describe the TD(λ) algorithm for learning the value function of a given policy.