PerfDojo: Automated ML Library Generation for Heterogeneous Architectures
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908631256006656 |
|---|---|
| author | Ivanov, Andrei Shen, Siyuan Gottardo, Gioele Chrapek, Marcin Boudaoud, Afif Schneider, Timo Benini, Luca Hoefler, Torsten |
| author_facet | Ivanov, Andrei Shen, Siyuan Gottardo, Gioele Chrapek, Marcin Boudaoud, Afif Schneider, Timo Benini, Luca Hoefler, Torsten |
| contents | The increasing complexity of machine learning models and the proliferation of diverse hardware architectures (CPUs, GPUs, accelerators) make achieving optimal performance a significant challenge. Heterogeneity in instruction sets, specialized kernel requirements for different data types and model features (e.g., sparsity, quantization), and architecture-specific optimizations complicate performance tuning. Manual optimization is resource-intensive, while existing automatic approaches often rely on complex hardware-specific heuristics and uninterpretable intermediate representations, hindering performance portability. We introduce PerfLLM, a novel automatic optimization methodology leveraging Large Language Models (LLMs) and Reinforcement Learning (RL). Central to this is PerfDojo, an environment framing optimization as an RL game using a human-readable, mathematically-inspired code representation that guarantees semantic validity through transformations. This allows effective optimization without prior hardware knowledge, facilitating both human analysis and RL agent training. We demonstrate PerfLLM's ability to achieve significant performance gains across diverse CPU (x86, Arm, RISC-V) and GPU architectures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_03586 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PerfDojo: Automated ML Library Generation for Heterogeneous Architectures Ivanov, Andrei Shen, Siyuan Gottardo, Gioele Chrapek, Marcin Boudaoud, Afif Schneider, Timo Benini, Luca Hoefler, Torsten Performance Artificial Intelligence The increasing complexity of machine learning models and the proliferation of diverse hardware architectures (CPUs, GPUs, accelerators) make achieving optimal performance a significant challenge. Heterogeneity in instruction sets, specialized kernel requirements for different data types and model features (e.g., sparsity, quantization), and architecture-specific optimizations complicate performance tuning. Manual optimization is resource-intensive, while existing automatic approaches often rely on complex hardware-specific heuristics and uninterpretable intermediate representations, hindering performance portability. We introduce PerfLLM, a novel automatic optimization methodology leveraging Large Language Models (LLMs) and Reinforcement Learning (RL). Central to this is PerfDojo, an environment framing optimization as an RL game using a human-readable, mathematically-inspired code representation that guarantees semantic validity through transformations. This allows effective optimization without prior hardware knowledge, facilitating both human analysis and RL agent training. We demonstrate PerfLLM's ability to achieve significant performance gains across diverse CPU (x86, Arm, RISC-V) and GPU architectures. |
| title | PerfDojo: Automated ML Library Generation for Heterogeneous Architectures |
| topic | Performance Artificial Intelligence |
| url | https://arxiv.org/abs/2511.03586 |