# 6.7960 Deep Learning, Fall 2026

https://deeplearning6-7960.github.io/

|   
### MIT EECS
6.7960 Deep Learning  |  
| --- |  
| 
### Fall 2026
 |  
|  [ [Schedule](https://deeplearning6-7960.github.io/#schedule) | [Policies](https://deeplearning6-7960.github.io/#collaboration_policy) | [Piazza](https://piazza.com/class/mtncdc6cx5o3p9) | [Canvas](https://canvas.mit.edu/courses/40032) | [Gradescope](https://www.gradescope.com/courses/1387009) | [Lecture Recordings](https://canvas.mit.edu/courses/40032/external_tools/6183) | [Office Hours](https://deeplearning6-7960.github.io/#office_hours) | [Previous years](https://deeplearning6-7960.github.io/#previous_years) ]  |  
  

## Course Overview
**Description** : Fundamentals of deep learning, including both theory and applications. Topics include neural net architectures (MLPs, CNNs, RNNs, transformers), geometry and invariances in deep learning, backpropagation and automatic differentiation, learning theory and generalization in high-dimensions, and applications to computer vision, natural language processing, and robotics. 
**Pre-requisites** : 18.05 and (6.3720, 6.3900, or 6.C01) 
**Note** : This course is appropriate for advanced undergraduates and graduate students, and is 3-0-9 units. Due to heavy enrollment, we will very unfortunately not be able to take cross-registrations this semester. 
Any and all personal or logistical questions, such as regarding absenses, accomodations, etc should be emailed to the course email, 6.7960-instructors@mit.edu, and not to the instructors directly. 
  

* * *
  

## Course Information
## Instructor [**Phillip Isola**](http://web.mit.edu/phillipi/)
phillipi at mit dot edu
OH: Mon 1-2 PM
## Instructor [**Kaiming He**](https://people.csail.mit.edu/kaiming/)
kaiming at mit dot edu
OH: Mon 9-10 AM
## Course Assistant [**Taylor Braun**](https://deeplearning6-7960.github.io/)
tvbraun at mit dot edu
## Head TA **Ning Zhang**
ningrz at mit dot edu
## Head TA **Lana Xu**
ylanaxu at mit dot edu
## TA **Aimee Yu**
aimeeyu at mit dot edu
## TA **Amit Schechter**
amitsch at mit dot edu
## TA **Amy Chang**
akchang at mit dot edu
## TA **Anakha Ganesh**
anakhag at mit dot edu
## TA **Aryan Bora**
abora at mit dot edu
## TA **Ashkan Soleymani**
ashkanso at mit dot edu
## TA **Cindy Wei**
cindywei at mit dot edu
## TA **David Baek**
dbaek at mit dot edu
## TA **Gabriel Manso**
gmanso at mit dot edu
## TA **Jake Austin**
j_austin at mit dot edu
## TA **Liam Sheldon**
lsheldon at mit dot edu
## TA **Maggie Lin**
maggiejl at mit dot edu
## TA **MingYang Deng**
dengm at mit dot edu
## TA **Rebecca Wang**
rebecca1 at mit dot edu
## TA **Shobhita Sundaram**
shobhita at mit dot edu
## TA **Suraj Reddy**
surajrdy at mit dot edu
## TA **Tina Wang**
txw at mit dot edu
## TA **Vanessa Xiao**
vzxiao at mit dot edu
## TA **Xingjian Bai**
xbai at mit dot edu
### - Logistics
  * Class meetings: Tuesday, Thursday 1:00 - 2:30 PM in room **45-230**. The overflow room is **34-101**.
  * We will be using both Piazza and Canvas for announcements.
  * Refer to: [Piazza](https://piazza.com/class/mtncdc6cx5o3p9) (all questions), [Canvas](https://canvas.mit.edu/courses/40032) (announcements), [Gradescope](https://www.gradescope.com/courses/1387009) (homework release, submission, and grades), and [Google Calendar](https://calendar.google.com/calendar/u/1?cid=MGI1YmE1YjM0ZTk4ZDBiMjM5ZTgwOTU3MDdiOTU0MDcyMWFiNTk1MDRlMTgwMmE2Zjk4NzBhZmMwZThjMjA0MkBncm91cC5jYWxlbmRhci5nb29nbGUuY29t) (office hours). 
  * **All extension requests must go through S3 or GradSupport. For any personal or logistical questions, such as regarding absenses, accomodations, etc, please email the course email, 6.7960-instructors@mit.edu, not the instructors directly.**


### - Grading Policy
* **Problem sets (20%)**
  * 5 psets, 4% each, each ~2 weeks long
  * Derivations, written responses, and coding
  * Grading will be pass/fail and done by an AI agent, which will give feedback, with oversight by TAs


* **Midterm Exam (30% + 5% for practice exam) and Final Exam (40% + 5% for practice exam)**
  * 2 hour exam, closed book, no computers, 1 double-sided page of handwritten notes
  * Practice exams will be provided
  * Practice exams will be graded pass/fail and given feedback by AI agents


* Regrade requests must come within 2 weeks of when we release each grade.
* [Collaboration policy](https://deeplearning6-7960.github.io/#collaboration_policy)
* [AI assistants policy](https://deeplearning6-7960.github.io/#AI_policy)
* [Attendance policy](https://deeplearning6-7960.github.io/#attendance_policy)
* [Late policy](https://deeplearning6-7960.github.io/#late_policy)
### - Materials
  * Readings will come from a variety of sources and will be posted on the schedule each week.
  * Some readings are derived from the course textbook, which can be found for free online: [Foundations of Computer Vision](https://visionbook.mit.edu/).
  * The best textbook devoted entirely to deep learning is probably [Understanding Deep Learning](https://udlbook.github.io/udlbook/), which is freely available online.
  * Content from 6.390 Intro to ML can also be found [here](https://introml.mit.edu/notes/?fbclid=PAQ0xDSwMxDVNleHRuA2FlbQIxMAABp0r1wjiBU7px9Kf6ziMGCn6NGB3GhTW-QhmDeMG5oCD9T6qAQW5ItdrbpohF_aem_jL3v0-a5F6jpmw5iOtA7Aw) for those who want to brush up on ML concepts.


* * *
  

## Class Schedule
  

** class schedule is subject to change **   
| **Date**  | **Topics**  | **Speaker**  | **Course Materials**  | **Assignments**  |  
| --- | --- | --- | --- | --- |  
|  Week 1  |  
| Thu 9/10  | Course overview, introduction to deep learning  | Kaiming He  |  [Slides](https://www.dropbox.com/scl/fi/yp830sv2usqcl7rxwxbzg/lec1_intro_2026.pdf?rlkey=6k3jli23u9rnu8rm7zp3cbpdv&dl=0)   
  
Optional Reading:  
[Notation for this course](https://visionbook.mit.edu/notations.html)  
[Neural Networks](https://visionbook.mit.edu/neural_nets.html)  |   |  
|  Week 2  |  
| Tue 9/15  | How to train a neural net  | Phillip Isola  |  [Slides](https://www.dropbox.com/scl/fi/5c9oozgif09t0kgc86un7/2_backprop_2026.pdf?rlkey=bzwhf4gx7vjn08seh16koetui&dl=0)  
  
Required Reading:  
[ Gradient-Based Learning](https://visionbook.mit.edu/gradient_descent.html)  
[ Backprop](https://visionbook.mit.edu/backpropagation.html)  
  
Optional Reading:  
[Old Optimizer, New Norm: An Anthology](https://arxiv.org/abs/2409.20325)  | [pset 1 out](https://canvas.mit.edu/courses/40032/assignments/495346)  |  
| Thu 9/17  | Approximation theory  | Phillip Isola  |  [Slides](https://www.dropbox.com/scl/fi/6uxq0ddqs81ezbxtoihrb/3_approximation_PI.pdf?rlkey=an9ifv4dn2p7aj16dcpcohpvd&dl=0)  
[Interactive Demo](https://deeplearning6-7960.github.io/materials/demos/approximation/index.html)  
  
Optional Reading:   
[Deep learning theory notes](https://mjt.cs.illinois.edu/dlt/index.pdf) sections 2 and 5 (this is written at a rather advanced level; try to get the intuitions rather than all the details)   |   |  
| Thu 9/17  | PyTorch Tutorial: 6:00-7:00 PM, 34-101  | Aryan Bora  |  [Pytorch Tutorial Colab](https://drive.google.com/file/d/1-qDf2aq7Ut2rbcqTrU1CTUCI7O5vguw4/view?usp=sharing)  |   |  
|  Week 3  |  
| Tue 9/22  | Architectures: ConvNets  | Kaiming He  |  [Slides](https://www.dropbox.com/scl/fi/nx26snzhymgwe0xiar2k1/lec4_convnet.pdf?rlkey=6v72nuhxqd42pxo7kh1780zji&st=a513bopq&dl=0)  
  
Required Reading:  
[CNNs](https://visionbook.mit.edu/convolutional_neural_nets.html)  
  
Optional Reading:  
[CS231n: Convolutional Neural Networks](https://cs231n.github.io/convolutional-networks/)  |   |  
| Tue 9/22  | Agentic Coding Tutorial: 6:00-7:00 PM, 32-123  | Anakha Ganesh  |  [GitHub](https://github.com/anakhag07/deep-learning-agentic-coding-tutorial)  
[Slides](https://deeplearning6-7960.github.io/materials/tutorials/6_7960_Agentic_Coding_Tutorial.pdf)  |   |  
| Thu 9/24  | Architectures: Sequence Modeling  | Kaiming He  |  [Slides](https://www.dropbox.com/scl/fi/wx2dm6u91tsrv5odz5dzq/lec5_seq.pdf?rlkey=dr50tv5k47j7274mb79g6xq1d&st=onrb3kos&dl=0)  
  
Optional Reading:  
[Understanding LSTM Networks](https://colah.github.io/posts/2015-08-Understanding-LSTMs/)  
[The Unreasonable Effectiveness of Recurrent Neural Networks](https://karpathy.github.io/2015/05/21/rnn-effectiveness/)  |   |  
| Thu 9/24  | Agentic Coding Tutorial: 6:00-7:00 PM, 32-123  | Suraj Reddy  |  [GitHub](https://github.com/anakhag07/deep-learning-agentic-coding-tutorial)  
[Slides](https://deeplearning6-7960.github.io/materials/tutorials/6_7960_Agentic_Coding_Tutorial.pdf)  |   |  
|  Week 4  |  
| Tue 9/29  | Architectures: Transformers  | Phillip Isola  |  [Slides](https://www.dropbox.com/scl/fi/5ysub5qnicurgds48bj8k/6_transformers_2026.pdf?rlkey=w8pjhwwvxn2kojll96m0vqiz7&dl=0)  
  
Required Reading:  
[Transformers](https://visionbook.mit.edu/transformers.html)  |  pset 1 due   
[pset 2 out](https://canvas.mit.edu/courses/40032/assignments/496978)  |  
| Thu 10/1  | Generalization Theory  | Phillip Isola  |  [Slides](https://www.dropbox.com/scl/fi/ic2jl6ccpt7z383bk37l8/7_generalization_2026.pdf?rlkey=cb4sz8rqub3uajk7ofsd6n9g9&dl=0)  
Optional Reading:  
[Deep learning requires rethinking generalization](https://arxiv.org/abs/1611.03530)  
[Deep learning not so mysterious or different](https://arxiv.org/abs/2503.02113)  
[Double descent](https://www.pnas.org/doi/10.1073/pnas.1903070116)  |   |  
|  Week 5  |  
| Tue 10/6  |  Going Deep with Neural Networks  | Kaiming He  |   |   |  
| Thu 10/8  |  Representation Learning Methods  | Kaiming He  |   |   |  
|  Week 6  |  
| Tue 10/13  | No class — Monday schedule  |   |   |  pset 2 due  
pset 3 out   |  
| Thu 10/15  |  Representation Learning: Weight-Space Geometry  | Phillip Isola  |   |   |  
|  Week 7  |  
| Tue 10/20  |  Representation Learning: Information Theory  | Kaiming He  |   |   |  
| Thu 10/22  |  Foundation models: pre-training  | Phillip Isola  |   |   |  
|  Week 8  |  
| Tue 10/27  |  Foundation models: scaling laws  | Phillip Isola  |   |  pset 3 due  
pset 4 out   |  
| Thu 10/29  | Midterm  |   |   |   |  
|  Week 9  |  
| Tue 11/3  |  Generative models: basics  | Kaiming He  |   |   |  
| Thu 11/5  |  Generative models: VAE and GAN  | Kaiming He  |   |   |  
|  Week 10  |  
| Tue 11/10  | Generative models: Diffusion and Flows  | Kaiming He  |   |  pset 4 due  
pset 5 out   |  
| Thu 11/12  | Generalization (OOD)  | Phillip Isola  |   |   |  
|  Week 11  |  
| Tue 11/17  | Transfer learning: Models and Data  | Phillip Isola  |   |   |  
| Thu 11/19  | Inference-time Algorithms  | Phillip Isola  |   |   |  
|  Week 12  |  
| Tue 11/24  | Guest Lecture 1  | TBD  |   |   |  
| Thu 11/26  | No class — Thanksgiving  |   |   |   |  
|  Week 13  |  
| Tue 12/1  | Evaluation  | Phillip Isola  |   | pset 5 due  |  
| Thu 12/3  | Applying Deep Learning to Your Problems  | Kaiming He  |   |   |  
|  Week 14  |  
| Tue 12/8  | Guest Lecture 2  | TBD  |   |   |  
| Thu 12/10  | Guest Lecture 3  | TBD  |   |   |  
|  Week 15  |  
| Tue 12/15  | Final Exam  |   |   |   |  
* * *
  

## Collaboration policy
  * Psets should be written up individually and should reflect your own individual work. However, you may discuss with your peers, TAs, and instructors.
  * You should not copy or share complete solutions or ask others if your answer is correct (in person or via piazza/canvas).
  * If you work with anyone on the pset (other than TAs and instructors), list their names at the top of the pset.


* * *
  

## AI assistants policy
  * Our policy for using AI assistants is _identical_ to our policy for using human assistants.
  * This is a deep learning class and you _should_ try out all the latest AI assistants (they are pretty much all using deep learning). It's very important to play with them to learn what they can do and what they can't do. That's a part of the content of this course.
  * Just like you can come to office hours and ask a human questions (about the lecture material, clarifications about pset questions, tips for getting started, etc), you are very welcome to do the same with AI assistants.
  * But: just like you are not allowed to ask an expert friend to do your homework for you, you also should not ask an expert AI. 
  * If it is ever unclear, just imagine the AI as a human and apply the same norm as you would with a human.
  * If you work with any AI on a pset, briefly describe which AI and how you used it at the top of the pset (a few sentences is enough).


* * *
  

## Attendance policy
* Attendance is at your discretion. Recordings will be released [here](https://canvas.mit.edu/courses/40032/external_tools/6183) right after each class.
* * *
  

## Late policy
  * Homeworks will not be accepted more than 7 days after the deadline.
  * The grade on a homework received n days after the deadline (n<=7) will be multiplied by (1-n/14). We will round up to units of full days; submitting 1 hour late counts as using 1 late day.
  * Ten penalty days will be automatically waived for each student.
  
For example, let's say a student perfectly solves the first three homeworks but submits the first homework 8 days late, the second homework six days late and the third homework five days late. Then the student will score zero on the first homework, 100% on the second homework (using up six late days) and 100% * (1-1/14) = 92.9% on the third homework (using up the last four remaining late days).  
  
The slack days are meant to be used for all the normal circumstances of life: being behind on work, forgetting the deadline, having a conference to attend, etc. We will not grant further extensions for these routine issues. For any extension request (i.e. serious medical issues or life events) please contact [S3](https://studentlife.mit.edu/s3) (for undergrads) or [GradSupport](https://oge.mit.edu/student-support-development/gradsupport/) (for grad students) and we will work with them to find a good solution. 
  * We will not be able to support course incompletes.
* * *
  

## Office Hours
Office hours are listed on the calendar below. You can also [add this calendar](https://calendar.google.com/calendar/u/1?cid=MGI1YmE1YjM0ZTk4ZDBiMjM5ZTgwOTU3MDdiOTU0MDcyMWFiNTk1MDRlMTgwMmE2Zjk4NzBhZmMwZThjMjA0MkBncm91cC5jYWxlbmRhci5nb29nbGUuY29t) to your own Google Calendar.   
  
If you're interested in speaking to the TAs about what they're working on, check out [this post](https://piazza.com/class/mtncdc6cx5o3p9/post/43) on Piazza! Feel free to come to office hours and ask about anything you're curious about! 
* * *
  

## Previous years
[Fall 2025](https://deeplearning6-7960.github.io/2025/index.html)  
[Fall 2024](https://phillipi.github.io/6.7960/)  
[Fall 2023](https://phillipi.github.io/6.s898/index.html)  
[Fall 2022](https://phillipi.github.io/6.s898/2022/index.html)  
[Fall 2021](https://phillipi.github.io/6.s898/2021/index.html)
* * *