Performance-Aligned LLMs for Generating Fast Code
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914775106060288 |
|---|---|
| author | Nichols, Daniel Polasam, Pranav Menon, Harshitha Marathe, Aniruddha Gamblin, Todd Bhatele, Abhinav |
| author_facet | Nichols, Daniel Polasam, Pranav Menon, Harshitha Marathe, Aniruddha Gamblin, Todd Bhatele, Abhinav |
| contents | Optimizing scientific software is a difficult task because codebases are often large and complex, and performance can depend upon several factors including the algorithm, its implementation, and hardware among others. Causes of poor performance can originate from disparate sources and be difficult to diagnose. Recent years have seen a multitude of work that use large language models (LLMs) to assist in software development tasks. However, these tools are trained to model the distribution of code as text, and are not specifically designed to understand performance aspects of code. In this work, we introduce a reinforcement learning based methodology to align the outputs of code LLMs with performance. This allows us to build upon the current code modeling capabilities of LLMs and extend them to generate better performing code. We demonstrate that our fine-tuned model improves the expected speedup of generated code over base models for a set of benchmark tasks from 0.9 to 1.6 for serial code and 1.9 to 4.5 for OpenMP code. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_18864 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Performance-Aligned LLMs for Generating Fast Code Nichols, Daniel Polasam, Pranav Menon, Harshitha Marathe, Aniruddha Gamblin, Todd Bhatele, Abhinav Distributed, Parallel, and Cluster Computing Artificial Intelligence Software Engineering Optimizing scientific software is a difficult task because codebases are often large and complex, and performance can depend upon several factors including the algorithm, its implementation, and hardware among others. Causes of poor performance can originate from disparate sources and be difficult to diagnose. Recent years have seen a multitude of work that use large language models (LLMs) to assist in software development tasks. However, these tools are trained to model the distribution of code as text, and are not specifically designed to understand performance aspects of code. In this work, we introduce a reinforcement learning based methodology to align the outputs of code LLMs with performance. This allows us to build upon the current code modeling capabilities of LLMs and extend them to generate better performing code. We demonstrate that our fine-tuned model improves the expected speedup of generated code over base models for a set of benchmark tasks from 0.9 to 1.6 for serial code and 1.9 to 4.5 for OpenMP code. |
| title | Performance-Aligned LLMs for Generating Fast Code |
| topic | Distributed, Parallel, and Cluster Computing Artificial Intelligence Software Engineering |
| url | https://arxiv.org/abs/2404.18864 |