Syllabus
Course goals, prerequisites, policies and grading structure.
Johns Hopkins University • Department of Computer Science
CS 475/675:
CS 475/675:
Machine Learning
A structured introduction to the principles, algorithms and mathematical foundations that allow computational systems to learn patterns, make predictions and improve from data.
Foundations
Probability, linear algebra and optimization
Algorithms
Regression, classification and graphical models
Practice
Assignments, experiments and a final project
Lecture Notes
Weekly topics, slides, reading references and review material.
Assignments
Analytical exercises, programming work and submission details.
Course Project
Milestones for proposing, developing and presenting an ML project.
Learn the theory. Build the models. Test the ideas.
Course Overview
From mathematical foundations to modern learning systems.
CS 475/675 introduces the core ideas of machine learning through a combination of theory, algorithm design and practical modeling. Students learn how to define learning problems, evaluate models and reason carefully about performance, uncertainty and generalization.
- Understand supervised, unsupervised and probabilistic learning methods
- Implement and evaluate machine learning algorithms using real data
- Analyze generalization, regularization, optimization and model assumptions
Core Topics
Open Course Resources
A complete path through foundational machine learning.
This module list is intentionally broad. Update the order and reading links to match the current instructor's approved syllabus.
Learning Foundations
Problem formulation, training and test data, empirical risk, evaluation metrics and generalization.
Weeks 1–2Linear Models
Linear regression, logistic regression, loss functions, regularization and feature representation.
Weeks 3–4Optimization
Gradient methods, stochastic optimization, convexity and practical model-training considerations.
Weeks 5–6Kernel Methods
Margin-based learning, support vector machines, kernels and nonlinear decision boundaries.
Weeks 7–8Probabilistic Models
Maximum likelihood, Bayesian reasoning, latent variables, expectation maximization and graphical models.
Weeks 9–11Modern Extensions
Ensembles, neural networks, structured prediction, representation learning and responsible ML.
Weeks 12–14
Sample Schedule
Week 01
Introduction
Week 02
Foundations
Week 03
Supervised ML
Week 04
Classification
Week 05
Optimization
Week 06+
Advanced Topics
Weekly progression through the course.
Dates and resources below are placeholders rather than an active academic calendar.
What Is Machine Learning?
Course organization, learning problems, datasets and evaluation.
Probability & Statistical Review
Random variables, conditional probability and maximum likelihood.
Linear Regression
Least squares, features, estimation and model evaluation.
Logistic Regression
Classification, likelihood, decision boundaries and regularization.
Gradient-Based Optimization
Gradient descent, stochastic methods and optimization diagnostics.
Kernels, Graphical Models & Neural Networks
Advanced methods, applications, project development and review.
Coursework
Analytical
Programming
Capstone
Assignments designed to connect analysis with implementation.
Replace the sample weights and deadlines with the current grading policy.
Problem Sets
Derivations and conceptual exercises covering probability, estimation, optimization and learning theory.
Machine Learning Labs
Implement models, run experiments and explain empirical results using Python-based workflows.
Final Project
Define a learning problem, select methods, evaluate outcomes and communicate conclusions clearly.
Student Resources
Everything students need in one organized place.
Lecture Slides
Link current slides, notes and supplementary readings by week.
Starter Code
Provide approved notebooks, datasets and assignment templates.
Discussion Board
Link the current official platform used for course questions.
Office Hours
Publish current staff availability, format and location details.
Course Staff
Instruction, guidance and support throughout the semester.
All staff profiles below are placeholders. Replace them with the current instructor and teaching-assistant information before publishing.
Instructor
Course Instructor
Add the current instructor's name, research interests, office location and email.
Teaching Assistant
Lead TA
Add recitation schedule, office hours, contact details and support responsibilities.
Course Assistant
Course Support
Add grading, logistics, discussion-board and assignment support information.
Frequently Asked Questions
Common questions before the course begins.
Update these answers to match the official current-term syllabus.
Official enrollment question?
Students should rely on the university catalogue, registrar and current course staff for authoritative information.
Students generally benefit from prior programming experience and familiarity with probability, statistics, linear algebra and basic algorithms.
The undergraduate and graduate registrations cover closely related material. Confirm any different requirements with the current instructor and official syllabus.
Python is commonly used for machine learning coursework, but students should verify the current assignment environment and permitted libraries.
Availability depends on the current course policy. Link only materials the instructor has approved for public or enrolled-student access.
Collaboration rules differ by assignment. Students must follow the current academic-integrity and attribution policy stated in the official syllabus.
Department Information
Stay connected with Johns Hopkins Computer Science.
Use official department and university systems for registration, policy questions and verified academic information.
- 📍 DepartmentJohns Hopkins Department of Computer Science, 3400 North Charles Street, Baltimore, MD 21218
- 🌐 Department websitewww.cs.jhu.edu
- 🎓 UniversityJohns Hopkins University
Template notice: This is a design template for a course-resource website.
It does not itself confirm current Johns Hopkins instructors, dates, policies, enrollment,
assignments or authorization. Replace all placeholders and secure appropriate university
approval before presenting the site as an official Johns Hopkins resource.