CSE 474/574: Introduction to Machine Learning - Spring 2024

Course Information

Instructor: Poonam Kumari M 4:00 – 6:00 212 Capen - poonamku@buffalo.edu

Teaching Assistants

Namit - W 4:00 – 5:00, F 2:00 – 3:00 212 Capen - namitjun@buffalo.edu
Mahesh – Th F 4:00 – 5:00 Bansal Atrium, Davis - mbhosale@buffalo.edu Zoom OH
Su – MT 11:00 – 12:00 Davis 106 - shaoshus@buffalo.edu

Time: Mon/Wed/Fri 3:00 - 3:50 p.m.

All lectures are live streamed and can be accessed using Brightspace

Piazza link: Signup

Assignments/exams will be accessible on piazza.

Course Overview

Machine learning (ML) is a subset of artificial intelligence (AI) that focuses on enabling computers to learn and make decisions or predictions from data without being explicitly programmed for every task. The goal of machine learning is to develop algorithms that can learn patterns or representations from data and generalize that knowledge to new, unseen data to perform tasks more accurately or efficiently. Three major goals of the course are:

To provide a broad survey of approaches and techniques in ML.
To understand the concept of the topics in depth along with the mathematical formulations designed to propose them.
To develop the design and implementation skills that will help you to build intelligent, adaptive artifacts.

Prerequisites

Python
Linear Algebra
Advanced Calculus
Statistics
CSE250: Data Structures
EAS 305 or MTH 411 or STA 301 or MTH 309

Tentative Schedule

Week	Date	Topic	Assignment	Project
1	1/24	Course Introduction
1	1/26	Linear Algebra
2	1/29	Probability Theory
	1/31	Probability Theory
	2/2	Introduction to Data Science
3	2/5	Linear Models
	2/7	Linear Models
	2/9	Linear Models
4	2/12	Linear Models	Assignment 1 Details
	2/14	Linear Models		Project 1 Details
	2/16	Linear Models	Assignment 1 submission
5	2/19	Non-Linear Models
	2/21	Non-Linear Models
	2/23	Non-Linear Models
6	2/26	Non-Linear Models	Assignment 2 details
	2/28	Non-Linear Models
	3/1	Non-Linear Models
7	3/4	Non-Linear Models
	3/6	Non-Linear Models
	3/8	Kernel Methods		Project 1 Submission/Project 2 Details
8	3/11	Project Discussion	Assignment 2 submission
	3/13	Kernel Methods
	3/15	Graphical Models
9	3/18	Spring Break
	3/20	Spring Break
	3/22	Spring Break
10	3/25	Graphical Models
	3/27	Graphical Models
	3/29	Graphical Models		Project 2 submission
11	4/1	Unsupervised Learning		Project 3 Details
	4/3	Unsupervised Learning
	4/5	Unsupervised Learning
12	4/8	No Class
	4/10	Project 3 info session
	4/12	Unsupervised Learning
13	4/15	Ensemble Methods
	4/17	Ensemble Methods
	4/19	Ensemble Methods
14	4/22	Hyperparameter Optimization
	4/24	Hyperparameter Optimization
	4/26	Hyperparameter Optimization		Project 3 submission
15	4/29	Ethical Issues of AI, Trustworthy & Explainable AI
	5/1	Ethical Issues of AI, Trustworthy & Explainable AI
	5/3	Practice Exam Discussion

Topics to be Covered

Introduction and Probability Overview
Linear Models (≈ 2.5 weeks): This will include Logistic Regression/Perceptrons, Support Vector Machine (SVM).
Non-linear Models (≈ 2 weeks): This will include Non-linear Regularization, Neural Network.
Kernel Methods (≈ 1 week):Multi Kernel Regression, Kernel SVM.
Graphical Models (≈ 2 weeks): Bayesian Networks, MRV, Inference in Graphical Models
Unsupervised Learning (≈ 1.5 weeks): Clustering, Dimension Reduction
Hyperparameter Optimization (≈ 1 week).
Ensemble Methods (≈ 1 week)
Ethical Issues of AI, Trustworthy & Explainable AI (≈ 1 week)

Textbooks

Pattern Recognition and Machine Learning, by Chris Bishop (2006).
Machine Learning: a Probabilistic Perspective, by Kevin Murphy (2012).
Introduction to Machine Learning, 3rd Edition, by Ethem Alpaydin (2015).

Grading Weights

3 Projects: 60% total
Assignments: 15% each
Final Exam: 25%

Late Policy

Late submissions will not be graded.

Academic Integrity

Team of 2 allowed on Assignments. Please register on Brightspace

Per departmental policy of academic integrity violations, if a violation is found, you will get F in the course and will be reported to the department AI committee.