COBAYN
Problem
Porting and tuning application code across diverse architectures is slow and costly, and standard compiler optimization levels often fail to deliver the best performance.
Why it matters
Compiler tuning time directly affects software performance and programmer productivity across platforms.
What the AI did
COBAYN uses Bayesian networks plus dynamic application characterization from independent microarchitecture features to learn compiler optimization choices, aiming to speed up applications and cut optimization cost.
Results
not stated in source
Impact
Published in a peer-reviewed ACM journal with 80 citations per the evidence.
Record
| Venue | ACM Transactions on Architecture and Code Optimization |
|---|---|
| Published | 2016 |
| Work type | article |
| Authors | Amir Hossein Ashouri, Giovanni Mariani, Gianluca Palermo, Eun Jung Park, John Cavazos, Cristina Silvano |
| Citations | 92 (OpenAlex, at import time) |
| References | 57 |
| Model or method | Bayesian networks |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1145/2928270 |
| Open access | bronze |
| Keywords | compiler autotuning, Bayesian networks, machine learning, performance optimization |
Importance rating
real but narrower advance
Compiler tuning affects software performance and programmer productivity across platforms; abstract reports speedups up to 4.6x on 39 applications on ARM/GCC.
More in Software & Coding
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | ImageNet classification with deep convolutional neural networks | 2017 | 106,345 |
8
|
| 2 | Deep Residual Learning for Image Recognition | 2016 | 221,134 |
8
|
| 3 | Faster sorting algorithms discovered using deep reinforcement learning | 2023 | 175 |
8
|
| 4 | A deep learning‐based approach for software vulnerability detection using code metrics | 2022 | 42 |
7
|
| 5 | Deep Learning-Based Attack Detection and Classification in Android Devices | 2023 | 41 |
7
|