EECS7398 Weekly Schedule fa 2024
Since the is the first time this course is offered.
This is a tentative schedule.
The papers listed here are suggestions, if there is a different paper you would like to present, send me a link so that I can approve it.
Either 1 or 2 people can sign up for a paper. Once everyone has done so, I’ll schedule which paper gets which date.
| session | Date | topic | due | discussions |
|---|---|---|---|---|
| 1 | Friday Sept 6 | Compiler Overview | discussion | |
| Performance Measurement-1 | ||||
| 2 | Tuesday Sept 10 | Performance Measurement-2 | discussion | |
| 3 | Friday Sept 13 | Representing programs | hw0 | discussion |
| Bril | discussion | |||
| 4 | Tuesday Sept 17 | Local analysis and optimization | discussion | |
| 5 | Friday Sept 20 | Local Value Numbering | hw1 | |
| 6 | Tuesday Sept 24 | lvn + data flow | ||
| 7 | Friday Sept 27 | Data flow continued | ||
| 8 | Tuesday Oct 1 | Data flow + graphs | ||
| 9 | Friday Oct 4 | Static single assignment | ||
| 10 | Tuesday Oct 8 | register allocation | ||
| 11 | Friday Oct 11 | Norm unavailable - guest lecture over zoom | hw3 |
| 12 | Tuesday Oct 15 | reading |
Leader: Aymane Jerari The MLIR Transform Dialect
| 13 | Friday Oct 18 | GPU Compilers | hw4 | |
| 14 | Tuesday Oct 22 | LLVM |
project proposal due
| 15 | Friday Oct 25 | reading |
Leader: Oscar Kellner Retargeting and Respecializing GPU Workloads for Performance Portability
| 16 | Tuesday Oct 29 | reading |
Leader: Sana Anvari Large Language Models for Compiler Optimization
| 17 | Friday Nov 1 | instruction scheduling | hw5 | |
| 18 | Tuesday Nov 5 | reading |
Leader: Qucheng Jiang An MLIR-based Compiler Flow for System-Level Design and Hardware Acceleration
| 19 | Friday Nov 8 | divergence | ||
| 20 | Tuesday Nov 12 | reading |
Leader: Yashaswini Makaram ProGraML: Graph-based Deep Learning for Program Optimization and Analysis
| 21 | Friday Nov 15 | Dynamic compilers 1 | ||
| 22 | Tuesday Nov 19 | reading |
Leaders: Rohit Anandakumar and Sharmila Sivalingam Generating GPU Compiler Heuristics using Reinforcement Learning
| 23 | Friday Nov 22 | Dynamic compilers | ||
| 24 | Tuesday Nov 26 | reading |
Leader: Michael Maurer Energy-Aware Tile Size Selection for Affine Programs on GPUs
| Nov 29 | Thanksgiving | |||
| 26 | Tuesday Dec 3 | reading |
Leader: Matin Raayai: Dr.Jit: A Just-In-Time Compiler for Differentiable Rendering
| 27 | Friday Dec 6 | ai in compilers |
Papers
extra topics
ra checking, add more about ai
extra papers
TVM: An Automated End-to-End Optimizing Compiler for Deep Learning](https://www.usenix.org/conference/osdi18/presentation/chen) paper
leader: SLaDe: A Portable Small Language Model Decompiler for Optimized Assembly
Leader:ACPO: AI-Enabled Compiler-Driven Program Optimization
leader: Adaptive Online Context-Sensitive Inlining
leader: Threaded Code Variations and Optimizations
leader:Learning Compiler Pass Orders using Coreset and Normalized Value Prediction
leader: Learning to Optimize Tensor Programs
leader: End-to-end Deep Learning of Optimization Heuristics
leader: Compiler Fuzzing through Deep Learning
Back to top