Program Overview

Introduction to AI & ML I

The course provides an integrated introduction to artificial intelligence and machine learning, bridging core AI methods with modern approaches. Students develop both theoretical mastery and practical expertise by combining foundational concepts with the construction of influential AI systems.

Curriculum

The curriculum covers foundational materials in search, machine learning, reinforcement learning, and probability. Students then build on these to construct detailed implementations of landmark AI systems such as AlexNet, GPT-2, and AlphaZero. This rigorous approach develops the analytical skills needed to build the future AI. Finally, as an essential component, this course will address the ethics and responsible development of AI/ML technology and products.

Course Emphasis

The course emphasizes both technical excellence and ethical considerations in AI development. It serves as the foundation for 07-380 Artificial Intelligence and Machine Learning II, which explores advanced topics, research methods, and specialized applications.

Comparison to 10-301

Both courses cover sufficient material for an intro machine learning course. 07-280 includes non-ML AI techniques, while 10-301 focuses only on ML, naturally reaching a few additional ML topics.

Course Requirements

• Prereqs:
  • 15-122
  • Probability
  • Linear Algebra
  • 15-151/Concepts
• Coreq:
  • Calc 2

Course Fulfillments

• Fulfills the Intro ML prereq for later ML (10-XXX) courses: check_circle
• Fulfills the 07-280 prereq for 07-380 AI/ML II: check_circle
• Fulfills the 07-280 requirement for the AI Major, Additional Major, and Minor: check_circle
• Fulfills the AI elective for the Computer Science Major and Additional Major: check_circle
• Fulfills the Intro ML requirement for the Stat/ML Major: check_circle
• Fulfills the Intro ML prereq for the ML Concentration Minor: check_circle
• Fulfills the Intro ML prereq for the 5th year ML Master's: check_circle

Course Topics

• ML fundamentals from decision trees to neural networks: check_circle
• Transformer networks and Large Language Models: check_circle
• Reinforcement Learning: check_circle
• Additional topics:
  • Heuristic Search
  • Adversarial Search
  • Constraint Satisfaction Problems
  • ML Parallelism/GPU Basics
  • Monte Carlo Tree Search

Why this Course?

The goal is to replace the older AI and ML courses, 15-281 and 10-315, with two sequenced courses, 07-280 and 07-380, covering the breadth and depth required by the AI majors, with the first of the two courses covering core AI and ML concepts for SCS students taking only one AI course, as well as anyone at CMU who wants a good technical introduction to the field.

Benefits of the Restructure

• Flexibility to grow two AI courses
  • Adapting topics
  • Building on first course in the second course
• Better single AI course for non-AI majors
  • First course as accessible as 15-281 is now
  • First course includes core ML topics in addition to AI breadth

Retirement of 15-281 and 10-315

No, 15-281 and 10-315 are being retired and will not be offered in the future.

Course Instructors

The new courses will be taught by a mix of faculty, primarily from the Machine Learning and Computer Science Departments.

Course Schedule

Both courses, 07-280 and 07-380, will be offered every semester (Fall and Spring), with 07-380 first being offered in Fall 2026.

Topics in 07-380 AI & ML II

07-380 is designed to be more flexible in its topics from semester to semester, adapting based on our faculty's best understanding of what additional/advanced AI/ML topics students need to learn, especially those graduating with a major/minor in AI. Potential topics include:

• Deeper AI/ML Ethics
• MAP
• ML Theory: PAC Learning
• PCA
• Clustering and K-means
• Ensemble Methods: Bagging and Boosting
• Recommender Systems
• Linear programming
• Integer programming
• Propositional Logic
• SAT
• Logical Agents
• Classical Planning
• Bayes' Nets: Representation
• Bayes' Nets: Inference
• Bayes' Nets: Sampling
• HMMs
• Game Theory: Equilibrium
• Game Theory: Social Choice
• Vision Transformers
• Variational Autoencoders
• Diffusion
• Text to Image Generation
• Distributed Deep Learning
• Optimization: RMS, Momentum, Stability
• RLHF and DPO

Prerequisites and Corequisites

Yes, the prerequisites and corequisites are strict requirements for enrollment in 07-280.