Homework Exercises

Due Feb. 15 @ 5pm***, students need to complete a subset of assigned exercises to pass:

• 20/ 28 points (cumulative) on conceptual online multiple choice questions

  • Each question is worth 2 points (so 10 / 14 questions)

  • Roughly two will be assigned per lecture

  • Wrong answers can be corrected and resubmitted online after a 30 minute delay for full credit - there is no limit on the number of attempted submissions

• Any 4 of the below applied homework exercises (out of 8 assigned, roughly one per lecture) at a passing level

• See course homework policies and how to submit

***Due to a university holiday on Feb 15th, the deadline for the applied exercises only will be extended to Feb 16 at noon.

• This extension is to ensure all students have access to the dropbox location in the Huang Engineering Center.

• The deadline for the conceptual exercises is still Feb 15th at 5pm - at that time the electronic submission system will close.

Conceptual Questions (online multiple choice)

Applied Exercises (submit any 4 from the list below, one will be added per lecture)

All homework exercises are taken from ISL textbook.

All data sets necessary for exercises can be found here.

• Lecture 1: Introduction

  • section 2.4 #9 (Auto dataset)

• Lectures 2 and 3: Unsupervised Learning

  • section 10.7: #10

    • See R tutorial (Lab 2: Clustering) in section 10.5.1 for a walk-though of clustering identical to this problem.

    • See R tutorial (Lab 1: Principal Component Analysis) in section 10.4 for PCA.

• Lecture 4: Linear Regression

  • section 3.7 #9 (a-c,e,f) (Auto dataset)

• Lecture 5: Cross-validation

  • section 5.4: #8 (a-e)

    • note: LOOCV stands for "leave-one-out cross-validation"

    • See R tutorials 5.3.1, 5.3.2, 5.3.3.

• Lecture 5: Regularization

  • section 6.8: # 9 (a-d)

    • See R tutorial in section 6.6 (Lab 2: Ridge Regression and the Lasso) for a step-by-step walk-through on how to complete this exercise.

• Lecture 6: SVM

  • section 9.7: #4

    • See R tutorials in sections 9.6.1 and section 9.6.2 for support vector classifier/machine.

    • For Matlab users, refer to documentation for svmtrain.

• Lecture 7: CART

  • section 8.4: # 9

    • The required dataset is found here (ISLR). See the R tutorial (Lab 2: Fitting Regression Trees) in section 8.3.2 for a step-by-step walk-though of the techniques used in this problem.

• Lecture 8: Random Forests

  • section 8.4: #8

    • The dataset required is found here.

    • See R tutorial (Lab 2: Fitting Regression Trees) in section 8.3.2 and (Lab 3: Bagging and Random Forests) in section 8.3.3 for a walk-though of the techniques used in this problem.

Homework policies and How to Submit can be found here