Public Syllabus Summary

Machine Learning (Fall 2026)

This dated summary follows the course syllabus revised October 9, 2026. Use the course website for current materials and Canvas for enrolled-student announcements, submissions, and grades. Announced adjustments take precedence over this summary.

Course Description

A mathematically rigorous introduction to statistical ML for advanced undergraduate and master’s students. Rather than surveying many algorithms superficially, the course uses representative models to explain why methods work, what assumptions they require, and when they fail. Homework, labs, and a final project connect theory to implementation.

Prerequisites listed in the syllabus are STOR 215 or STOR 315 or MATH 381, and STOR 435 or STOR 535. Students should be prepared to use basic Python and read mathematical notation.

Course Structure and Goals

  1. Foundations: learning paradigms, empirical risk minimization, generalization, PAC learning, VC dimension, validation, and optimization.
  2. Predictive modeling and regularization: linear/logistic regression, stochastic gradients, regularization, support vector machines, margins, and kernels.
  3. Flexible prediction and structure discovery: nearest neighbors, decision trees, generative classification, clustering, PCA, and matrix factorization.
  4. Learned representations and generative AI: embeddings, representation objectives, language modeling, and the interpretation and deployment of generative models.

Students should be able to formulate learning problems; explain the role of assumptions; derive, implement, and analyze representative models; diagnose overfitting; design small Python experiments; and communicate results and limitations clearly.

Materials and Purchases

No required purchases or textbook. Slides, readings, and other materials are supplied through course materials or free public sources. Students need Python, Canvas, and Overleaf. Overleaf is required for homework and written reports; laptops are permitted and may be needed for computational work.

References include An Introduction to Statistical Learning, The Elements of Statistical Learning, and Foundations of Machine Learning. No single textbook is required.

Assessment and Participation

ComponentWeight
Four homework assignments20%, 5% each
Exam component50%
Final project30%: presentation 10%, report 20%

For mock-exam score M and final-exam score E, the exam component is max(0.40M + 0.60E, E). If the mock exam is missed, the final-exam score alone is used. The instructor reserves the right to curve the final score when necessary.

Attendance is not required. The instructor may award in-class bonus points for active, respectful engagement; absences can mean missing bonus, lab, or project opportunities.

Homework is submitted electronically as a PDF, not handwritten or scanned copies. The final project is completed in groups of four or five, with an in-class presentation and written report.

GradePercentage
A / A-93 or above / 90-92.99
B+ / B / B-87-89.99 / 83-86.99 / 80-82.99
C+ / C / C-77-79.99 / 73-76.99 / 70-72.99
D+ / D / D- / F67-69.99 / 63-66.99 / 60-62.99 / below 60

Course Policies

Collaboration and integrity. Collaboration on homework and projects is encouraged unless an assignment states otherwise. Students are bound by the Honor Code and should ask for clarification when uncertain.

Generative AI. GenAI/LLM tools are permitted as learning, drafting, coding, and problem-solving aids. Students must be able to explain and reproduce their submitted claims, derivations, citations, code, and conclusions independently. Unverified or fabricated AI output submitted as fact results in a zero for that assignment. Disclose when and how tools were used; disclosure itself does not lower the grade. GenAI and other external tools are prohibited during in-class exams unless explicitly authorized.

Late work. Homework loses 20% for each day late, with any portion counted as a full day. Missing a presentation earns zero for that presentation. A late final-project submission earns zero for all group members. Exam accommodations follow the syllabus and UNC’s official process.

Re-evaluation. A requested complete re-evaluation may raise, preserve, or lower the grade.

Email. Begin the subject line with STOR565. A response is not guaranteed within 24 hours of a deadline; regular inquiries should normally receive a response within 48 hours, excluding non-working time.

Diverse scholarly perspectives. This course engages diverse scholarly perspectives to develop critical thinking, analysis, and debate. Inclusion of a reading, case, argument, or other course material does not imply endorsement by the instructor or the University.

Dated Calendar Anchors

The October 9 syllabus schedules the mock exam for October 28 and the in-class final exam for November 16, with no additional exam in the University final-examination period. Final presentations are planned for November 18, 20, 23, 30, and December 2. The written-report deadline and submission instructions will be announced through Canvas. These are dated syllabus plans, not a check of subsequent announcements.

The syllabus is a guide, not an express or implied contract, and may change with advance announcements through Canvas. For accommodations and student support, consult the syllabus and the relevant UNC resources.

docs