APM496H

 

Instructor: Prof. Luis Seco, Director Master in Mathematical Finance (MMF),

Professor of Mathematics.

 

 

Open for third and fourth year undergraduate students with a good science or engineering background.

 

The course will consist of three different components:

 

1.     A general discussion on Data Science, ML/AI taught by Prof. Seco.

2.     A course on optimization and ML/AI by Prof. Roy Kwon

3.     Satellite discussions by subject matter experts:

a.     Nicholas Hoell, MMF and Data Scientist at Deloitte

b.     Rahul Raina, Microsoft

c.     Prof. Lennon Li, School of Public Health, University of Toronto

4.     Laboratory assignments, in Python, by University of Toronto Teaching Assistants:

a.     Yichao Chen

b.     Amin Sammara

 

APM496H Contents

 

Data Analysis

Exploratory Data Analysis

Model Formulation

Goodness of Fit Testing

Standard and Non-standard Statistical Analysis

Linear and non- linear Regression

Analysis of Variance

Time-series Analysis

 

Machine Learning

Feasibility of learning

Measures of Fit and Lift

Logistic Regression

Neural Networks

Support Vector Machines

Boosting, Decision Trees

 

Optimization

Linear and quadratic programming

Applications

 

Inference by Prof. Lennon Li

·      Part I: Introduction to biostatistics

·      Part II: inference

 

Lab and tutorials

Introduction to Programming for Data Science

 

Assignment 1

Assignment 2

Assignment 3

 

Prof. Seco’s Presentations:

 

 

 

Presentation by Rahul Raina, Microsoft.

 

Prof. Roy Kwon Lecture notes

 

March 9, Linear Algebra review

March 9, Optimization in ML

 

March 18:

·      Gradient Methods

·      Hyperparameter Cros Validation

·      Newtons for Optimization

·      One Dimensional Newton’s method

 

 

Course evaluation

 

The course will have two requirements that will be the basis for students marks:

1.     In-class participation (20%)

2.     Assignments (40%)

3.     Final Presentation