MIT · Course
6.S191: Introduction to Deep Learning
MIT 6.S191 is MIT’s introductory deep learning programme, described on its site as a high-intensity bootcamp. Its online edition releases new lectures, slides and labs openly each week, with applications to language, vision and biology.
Who it suits
- People who want the fundamentals of deep learning quickly: the site describes an efficient, high-intensity bootcamp.
- Listeners are welcome, and experience in Python is helpful but not necessary.
What it covers
- Foundational knowledge of deep learning algorithms and practical experience building neural networks.
- Applications to natural language processing, computer vision and biology.
- Recent topics, including large language models and generative AI.
- A project proposal competition with feedback from staff and industry sponsors, for those taking the in-person programme.
Watch here
From MIT’s own channel, so you can start without leaving Infrato. Nothing loads from YouTube until you press play.
Alexander Amini · 2026
MIT 6.S191: Reinforcement Learning
Alexander Amini · 2026
MIT Introduction to Deep Learning | 6.S191
Alexander Amini · 2026
MIT 6.S191: Deep Generative Modeling
Alexander Amini · 2026
MIT 6.S191: Secrets of Massively Parallel Training
Alexander Amini · 2026
MIT 6.S191: The Three Laws of AI
Alexander Amini · 2026
MIT 6.S191: Recurrent Neural Networks, Transformers, and Attention
Alexander Amini · 2026
MIT 6.S191: Language Models and New Frontiers
Alexander Amini · 2026
MIT 6.S191: Convolutional Neural Networks
Alexander Amini · 2026
MIT 6.S191: AI for Science
Before you start
- Calculus, meaning taking derivatives, and linear algebra, meaning matrix multiplication.
- Python experience helps but is not required.
Common questions about 6.S191: Introduction to Deep Learning
- Is 6.S191 free to follow?
- The site says the course is open-sourced to the world and that its online edition releases new lectures, slides and labs every week. Check the site for the current edition.
- What maths do I need?
- The site says it assumes calculus (taking derivatives) and linear algebra (matrix multiplication) and explains the rest along the way.
- Do I need to know Python?
- Experience in Python is helpful but not necessary, according to the course site.
- Which edition did we read?
- The 2026 edition: its in-person week was held in January 2026, and the online edition releases weekly.
More like this
Keep learning on Infrato
- Path: AI for software developers
- Path: AI for data analysts
- More from MIT: all MIT listings in the directory
- Explore all AI courses
Where this comes from
Written from MIT’s own pages, read on 2026-10-13. Content, prices and availability change, so the provider’s page is the source. We are not affiliated with MIT and this is not an endorsement.