The best free CMU AI courses, curated
Carnegie Mellon runs one of the deepest AI and machine-learning programs in the world, and several of its flagship courses put full lecture recordings and slide decks online for free. Below are the standouts — deep learning, NLP and multimodal ML — organized into a path with our notes on where to start.
These are curated links to publicly available course materials hosted by Carnegie Mellon University and its instructors (course sites and YouTube). All content belongs to its respective authors. AnybodyCanAI is not affiliated with, sponsored by, or endorsed by CMU. The CMU name is used only to identify the source of each course.
Introduction to Deep Learning
From the perceptron to modern deep nets
Bhiksha Raj, Rita Singh · Fall 2024 · Language Technologies Institute
Summary
A rigorous, full-semester tour of deep learning that starts at the perceptron and builds up through optimization, convolutional and recurrent networks, attention and transformers — with the mathematical grounding CMU is known for.
One of the most complete deep-learning courses on the open internet: every lecture is recorded on YouTube and every slide deck is a public PDF. If you want depth and theory behind the practice, this is the one.
Schedule
Lecture topics and dates rotate each offering — we link straight to Carnegie Mellon University's official schedule and syllabus so it's always current.
Full schedule & syllabusAccess
Lecture recordings are on the course YouTube channel and slide decks are public PDFs on the course site. Graded homeworks/Kaggle competitions are for enrolled students, but the materials are free to follow. Strong linear algebra and Python help.
Materials & links
Source: Carnegie Mellon University — all links open on the provider's own site.
Advanced NLP
Modern NLP, from fundamentals to research
Graham Neubig · Fall 2024 · Language Technologies Institute
Summary
A graduate-level course on modern natural language processing: neural network fundamentals, language modeling, pretraining and fine-tuning, prompting, retrieval and augmentation, agents, and how to actually evaluate NLP systems — taught by a leading NLP researcher.
Graham Neubig keeps this course on the frontier and posts the full schedule, slides and lecture videos publicly. It is one of the best ways to go from "I use LLMs" to understanding how they are built, trained and measured.
Schedule
Lecture topics and dates rotate each offering — we link straight to Carnegie Mellon University's official schedule and syllabus so it's always current.
Full schedule & syllabusAccess
The course site publishes the full schedule and slide decks; lecture recordings are posted on the instructor's YouTube. Assignments are geared to enrolled students but the materials are free to follow.
Materials & links
Source: Carnegie Mellon University — all links open on the provider's own site.
Multimodal Machine Learning
Learning across language, vision and beyond
Louis-Philippe Morency, Paul Liang · Fall 2023 · Language Technologies Institute
Summary
A structured course on multimodal machine learning — how models represent, align and fuse information across language, vision, audio and more — built around a well-known taxonomy of the core technical challenges in the field.
Multimodal models (text + image + audio) are where a lot of frontier AI is heading, and this is one of the few free, rigorous courses dedicated entirely to it. The lectures and reading list are a superb map of the area.
Schedule
Lecture topics and dates rotate each offering — we link straight to Carnegie Mellon University's official schedule and syllabus so it's always current.
Full schedule & syllabusAccess
The course site publishes the syllabus, schedule and reading list; lecture recordings are on the group's YouTube channel. Free to follow.
Materials & links
Source: Carnegie Mellon University — all links open on the provider's own site.