Metraipolis › archive › dc_8e71ff76439e

Energy-Efficient Reinforcement Learning for Large Language Models — arXiv:2611.05748

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.05748
License (as archived)
CC0-1.0
Archived
2026-11-05T18:19:19.316Z
SHA-256 of the archived text
f1ec811d936e5a832ee50e19b47494f0902c2a163da82394709df5422f9e0bc4
Stable URL of this page
https://metraipolis.com/documento/dc_8e71ff76439e

How to cite

Energy-Efficient Reinforcement Learning for Large Language Models — arXiv:2611.05748 — arXiv (en), https://arxiv.org/abs/2611.05748 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8e71ff76439e (Metraipolis archive dc_8e71ff76439e, sha256 f1ec811d936e5a832ee50e19b47494f0902c2a163da82394709df5422f9e0bc4)
BibTeX
@misc{dc8e71ff76439e,
  title = {Energy-Efficient Reinforcement Learning for Large Language Models — arXiv:2611.05748},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.05748},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8e71ff76439e; sha256 f1ec811d936e5a832ee50e19b47494f0902c2a163da82394709df5422f9e0bc4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.05748. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reinforcement learning for language models alternates between generating responses and training the policy. These stages place different demands on GPUs, creating opportunities to reduce clock frequency with little loss in throughput. Selecting frequencies online is difficult because response lengths change and frequency trials themselves consume time and energy. We present EE-RL (energy-efficient RL), an online controller that selects separate frequencies for generation and training during RL execution. It compares candidate settings with a prediction of the default GPU governor on the same token workload, and keeps the most energy-efficient setting that meets a throughput threshold. The controller then monitors throughput and energy and adjusts frequencies as the workload changes. Fixed-frequency sweeps on H100 and GB200 show GPU energy savings of up to 28.2% per valid training token across synchronous and asynchronous execution, while largely preserving throughput. EE-RL turns this headroom into online savings, reducing GPU energy per valid training token by up to 16.9% at a modest throughput cost.

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.DC, nuovi lavori del giorno).
Categorie arXiv: Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2611.05748 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-05T18:13:18.986Z dalla pagina https://arxiv.org/list/cs.DC/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).