18.337 - Parallel Computing and Scientific Machine Learning
Really nice Julia tutorial for the fall 2022 class Tutorial
Julia tutorial by Steven Johnson Wed Feb 8 Optional Julia Tutorial: Wed Feb 8 @ 5pm via Zoom
Virtually via Zoom. Recording will be posted.
A basic overview of the Julia programming environment for numerical computations that we will use in 18.06 for simple computational exploration. This (Zoom-based) tutorial will cover what Julia is and the basics of interaction, scalar/vector/matrix arithmetic, and plotting — we'll be using it as just a "fancy calculator" and no "real programming" will be required.
If possible, try to install Julia on your laptop beforehand using the instructions at the above link. Failing that, you can run Julia in the cloud (see instructions above).
There will be homeworks, followed by the final project. Everyone needs to present their work and submit a project report.
1-page Final Project proposal due : March 24
Final Project presentations : April 26 to May 15
Final Project reports due: May 15
50% problem sets, 10% for the final project proposal, and 40% for the final project. Problem sets and final projects will be submitted electronically.
# | Notebook |
---|---|
1 | HW1 |
(For matrix calculus problems, do not use indices) | |
2 | HW2 Due Wednesday March 1, 2023 |
3 | HW3 Due Wednesday March 15, 2023 |
4 | HW4 Due Wednesday April 19, 2023 |
# | Day | Date | Topic | SciML lecture | Materials |
---|---|---|---|---|---|
1 | M | 2/6 | Intro to Julia. My Two Favorite Notebooks. | [Julia is fast], [AutoDiff], [autodiff video], | |
2 | W | 2/8 | Matrix Calculus I and The Parallel Dream | See [IAP 2023 Class on Matrix Calculus],[handwritten notes],[The Parallel Dream] | |
3 | M | 2/13 | Matrix Calculus II | [handwritten notes],[Corgi in the Washing Machine],[2x2 Matrix Jacobians] | |
4 | W | 2/15 | Serial Performance | 2 | [handwritten notes], [Serial Performance .jl file], [Loop Fusion Blog ] |
5 | T | 2/21 | Intro to PINNs and Automatic differentiation I : Forward mode AD | 3 and 8 | ode and Pinns,intro to pinn handwritten notes,autodiff handwritten notes |
6 | W | 2/22 | Automatic differentiation II : Reverse mode AD | 10 | pinn.jl, reverse mode ad demo,handwritten notes |
7 | M | 2/27 | Dynamical Systems & Serial Performance on Iterations | 4 | Lorenz many ways, Dynamical Systems, handwriten notes |
8 | W | 3/1 | HPC & Threading | 5 and 6 | pi.jl, threads.jl,HPC Slides |
9 | M | 3/6 | Parallelism | Parallelism in Julia Slides,reduce/prefix notebook | |
10 | W | 3/8 | Prefix (and more) | ppt slides, reduce/prefix notebook,ThreadedScans.jl,cuda blog | |
11 | M | 3/13 | Adjoint Method Example | 10 | Handwritten Notes |
12 | W | 3/15 | Guest Lecture - Chris Rackauckas | ||
13 | M | 3/21 | Vectors, Operators and Adjoints | Handwritten Notes | |
14 | W | 3/23 | Adjoints of Linear, Nonlinear, Ode | 11 | Handwritten Notes, 18.335 adjoint notes (Johnson) |
Spring Break | |||||
15 | M | 4/3 | Guest Lecture, Billy Moses | Enzyme AD | |
16 | W | 4/5 | Guest Lecture, Keaton Burns | Dedalus PDE Solver | |
17 | M | 4/10 | Adjoints of ODE | Handwritten Notes | |
18 | W | 4/12 | Partitioning | ||
M | 4/17 | Patriots' Day | |||
19 | W | 4/19 | Fast Multipole and Parallel Prefix | Unfinished Draft | |
20 | M | 4/24 | |||
21 | W | 4/26 | Project Presentation I | ||
22 | M | 5/1 | Project Presentation II | ||
23 | W | 5/3 | Project Presentation III | ||
24 | M | 5/8 | Project Presentation IV | ||
25 | W | 5/10 | Project Presentation V | ||
M | 5/15 | Class Cancelled |
|8|W|3/1| GPU Parallelism I |7| [video 1],[video2]
|9|M|3/6| GPU Paralellism II | | [video], [Eig&SVD derivatives notebooks], [2022 IAP Class Matrix Calculus]
|10|W|3/8| MPI | | Slides, [video, Lauren Milichen],[Performance Metrics] see p317,15.6
|11|M|3/13| Differential Equations I | 9|
|12|W|3/15| Differential Equations II |10 |
|13|M|3/20| Neural ODE |11 |
|14|W|3/22| |13 |
| | | | Spring Break |
|15|M|4/3| | | GPU Slides Prefix Materials
|16|W|4/5| Convolutions and PDEs | 14 |
|17|M|4/10| Chris R on ode adjoints, PRAM Model |11 | [video]|
|18|W|4/12| Linear and Nonlinear System Adjoints | 11 | [video]|
| |M|4/17| Patriots' Day
|19|W|4/19| Lagrange Multipliers, Spectral Partitioning || Partitioning Slides| |
|20|M|4/24| |15| [video],notes on adjoint|
|21|W|4/26| Project Presentation I |
|22|M|5/1| Project Presentation II | Materials
|23|W|5/3| Project Presentation III | 16 | [video]
|24|M|5/8| Project Presentation IV |
|25|W|5/10| Project Presentation V |
|26|M|5/15| Project Presentation VI|
Setting the stage for this course which will involve high performance computing, mathematics, and scientific machine learning, we looked at two introductory notebooks. The first [Julia is fast]](https://github.com/mitmath/18337/blob/master/lecture1/Julia%20is%20fast.ipynb) primarily reveals just how much performance languages like Python can leave on the table. Many people don't compare languages, so they are unlikely to be aware. The second [AutoDiff]](https://github.com/mitmath/18337/blob/master/lecture1/AutoDiff.ipynb) reveals the "magic" of forward mode autodifferentiation showing how a compiler can "rewrite" a program through the use of software overloading and still maintain performance. This is a whole new way to see calculus, not the way you learned it in a first year class, and not finite differences either.
We gave an example [The Parallel Dream]](https://github.com/mitmath/18337/blob/master/lecture1/the_dream.ipynb)
HW1 will be due Thursday Feb 16. This is really just a getting started homework.
For the second half of the class students will work on the final project. A one-page final project proposal must be sumbitted by March 24 Friday, through canvas.
Last three weeks (tentative) will be student presentations.
Here's a list of current projects of interest to the julialab
One possibility is to review an interesting algorithm not covered in the course and develop a high performance implementation. Some examples include:
Another possibility is to work on state-of-the-art performance engineering. This would be implementing a new auto-parallelization or performance enhancement. For these types of projects, implementing an application for benchmarking is not required, and one can instead benchmark the effects on already existing code to find cases where it is beneficial (or leads to performance regressions). Possible examples are:
pmapreduce
extension to pmap
that adds a paralellized reduction, or a fast GPU-based map-reduce.Additionally, Scientific Machine Learning is a wide open field with lots of low hanging fruit. Instead of a review, a suitable research project can be used for chosen for the final project. Possibilities include: