Section outline

  • Lecture: What is Machine Learning? Introductory concepts

    Supervised vs. Unsupervised learning. Overview of prediction techniques for supervised learning. Discussion of overfitting, loss functions, parametric vs. non-parametric models, and the bias-variance trade-off.

    Location: MH-1400

    Readings:

    • ISLR: Sec 2-2.2.2, Sec 3-3.2
    • ESL: Sec 1

    References for R:

    • https://class.lambdamd.org/pdsr/

    References for basic linear algebra:

    • Session Slides:

    • Session Audio/Video Recording (access restricted to registered students):

    • Lecture 1 Video Media Resource
      Not available unless: You belong to Registered Students Only
    • Homework

    • Homework 1 turn in Assignment

      Due 11:59 PM January 19th, 2022

      Not available unless: You belong to Registered Students Only
    • HW1 soln File
      Not available unless: You belong to Registered Students Only
  • Lab

    • Lab1 solutions File
      Not available unless: You belong to Registered Students Only
  • Lecture: Model fitting by optimization & Classification

    Linear regression as an optimization procedure; Introduction to classification models: logistic regression, multinomial regression, linear discriminant analysis, K-nearest neighbors, receiver operating characteristic (ROC)

    Readings:

    • ISLR: Sec 4-4.4.3
    • ESL: Sec 4 - 4.4

    • Session Slides:

    • Session Audio/Video Recording (access restricted to registered students):

    • Lecture 2 Media Resource
      Not available unless: You belong to Registered Students Only
  • Lab
    • Lab 2 solutions File
      Not available unless: You belong to Registered Students Only
  • Lecture: Penalized Regression and Classification

    Model selection by cross-validation; Low versus high-dimensional data; Regularization methods: Ridge, variable subset selection, Lasso


    ISLR: Sections 6.1-6.2, 6.4

    ESL: Sections 3.3-3.4

    • Session Slides:

    • Session Audio/Video Recording (access restricted to registered students):

    • lecture 3 Media Resource
      Not available unless: You belong to Registered Students Only
    • Homework

    • Due 11:59PM February 2nd, 2022

    • Homework 2 turn in Assignment

      Please turn in RmD and html file

      Not available unless: You belong to Registered Students Only
    • HW2 soln File
      Not available unless: You belong to Registered Students Only
  • Lab
    • Lab 3 solutions File
      Not available unless: You belong to Registered Students Only
  • Lecture:  Support Vector Machines
    Maximum margin classifiers, support vector classifiers, support vector machines

    ISLR: Sec 9

    ESL: Sec 4.5, 12.1-12.3

    • Session Slides:

    • Session Audio/Video Recording (access restricted to registered students):

    • Lecture 4 Media Resource
      Not available unless: You belong to Registered Students Only
  • Lab

    • Lab 4 solutions File
      Not available unless: You belong to Registered Students Only
  • Lecture: Tree-based methods

    Classification and regression trees; Bagging; Random forests

    ISLR: Sec 8
    ESL: Sec 9.2, 15
    • Session Slides:

    • Session Audio/Video Recording (access restricted to registered students):

    • Lecture 5 Media Resource
      Not available unless: You belong to Registered Students Only
    • Homework

    • Due 11:59PM, February 16th, 2022

    • Homework 3 turn in Assignment
      Not available unless: You belong to Registered Students Only
    • HW3 soln File
      Not available unless: You belong to Registered Students Only
  • Lab

    • lab5 solutions File
      Not available unless: You belong to Registered Students Only
  • Lecture: Boosting and gradient boosting
    Boosting, Adaboost, Gradient boosted trees

    ISLR: Sec 8.2.3

    ESL: Sec 10

    • Session Slides:

    • Session Audio/Video Recording (access restricted to registered students):

    • Lecture 6 Media Resource
      Not available unless: You belong to Registered Students Only
  • Lab


    • lab6 solutions File
      Not available unless: You belong to Registered Students Only
  • Lecture:  Unsupervised learning 

    Unsupervised learning: clustering and dimension reduction

    ISLR: Sec 10

    ESL: Sec 14.3, 14.5.1

    • Session Slides:

    • Session Audio/Video Recording (access restricted to registered students):

    • Lecture 7 Media Resource
      Not available unless: You belong to Registered Students Only
    • Homework

    • Due 11:59PM, March 2nd, 2022

    • Homework 4 turn in Assignment
      Not available unless: You belong to Registered Students Only
    • HW4 soln File
      Not available unless: You belong to Registered Students Only
  • Lab
  • Lecture: Neural networks

    Deep learning, Dense neural networks, Convolutional neural networks
    • Session Slides:

    • Session Audio/Video Recording (access restricted to registered students):

    • Lecture 8 Media Resource
      Not available unless: You belong to Registered Students Only
  • Lab

    • lab8 solutions with pytorch File
      Not available unless: You belong to Registered Students Only
  • Lecture:  Applications lecture: clustering and classification of high-dimensional data in a genomic context
    Clustering and classification of high-dimensional data in a genomic context

    Guest lecture by Adam Olshen

    • Session Slides:

    • Session Audio/Video Recording (access restricted to registered students):

    • Lecture 9 URL
      Not available unless: You belong to a group in Registered Students Only
    • Homework

    • Homework 5 turn in Assignment

      Due 11:59PM, March 16th, 2022

      Not available unless: You belong to Registered Students Only
  • Lab: We will discuss how to critique papers analyzing biomedical data using ML. 
    • Reading assignment 1: Molecular signatures from omics data: From chaos to consensus File
      Not available unless: You belong to Registered Students Only
    • Reading assignment 2: Minimum information about clinical artificial intelligence modeling: the MI-CLAIM checklist File
      Not available unless: You belong to a group in Registered Students Only
  • Lecture 10: Updating clinical prediction models

    Readings: Steyerberg "Clinical Prediction Models" Chapter 20

    • Session Slides:

    • Session Audio/Video Recording (access restricted to registered students):

    • Lecture 10 URL
      Not available unless: You belong to a group in Registered Students Only
  • Lab


    • Lab 10 solutions File
      Not available unless: You belong to Registered Students Only