Metraipolis › archive › dc_7d97f2caa65f
Turbo-GMM: FlashAttention-Inspired GPU-Accelerated Gaussian Mixture Model — arXiv:2611.02449
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02449
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T10:58:21.397Z
- SHA-256 of the archived text
68a488513697049ac43ba0a3a956e122e48a9073f991f2f81990f9380271f345- Stable URL of this page
https://metraipolis.com/documento/dc_7d97f2caa65f
How to cite
Turbo-GMM: FlashAttention-Inspired GPU-Accelerated Gaussian Mixture Model — arXiv:2611.02449 — arXiv (en), https://arxiv.org/abs/2611.02449 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7d97f2caa65f (Metraipolis archive dc_7d97f2caa65f, sha256 68a488513697049ac43ba0a3a956e122e48a9073f991f2f81990f9380271f345)
BibTeX
@misc{dc7d97f2caa65f,
title = {Turbo-GMM: FlashAttention-Inspired GPU-Accelerated Gaussian Mixture Model — arXiv:2611.02449},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02449},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7d97f2caa65f; sha256 68a488513697049ac43ba0a3a956e122e48a9073f991f2f81990f9380271f345},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02449. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Expectation-maximization (EM) for Gaussian mixture models (GMMs) requires scoring $N$ data points against $K$ mixture components. Prior fused GPU implementations, notably Flash-GMM, eliminate materialization of the $N \times K$ responsibility matrix but traverse the component dimension twice to compute normalization terms and accumulate sufficient statistics. We introduce Turbo-GMM, a single-sweep fused EM kernel that completes responsibility normalization and sufficient-statistic accumulation within one traversal of the component dimension. By retaining the complete component axis on chip, Turbo-GMM avoids the second score computation required by the released two-sweep Flash-GMM implementation. Its applicability is limited by register pressure, occupancy, and the component and feature dimensions, making it particularly suitable for large datasets with relatively small numbers of components. For larger configurations, Turbo-GMM employs a two-pass fallback following the established fused design. We establish algebraic equivalence to standard EM in real arithmetic and evaluate sixteen configurations on an NVIDIA A100. Under the reported benchmarking protocol, Turbo-GMM achieves a median $2.15\times$ speedup over the evaluated Flash-GMM implementation and up to $5.3\times$ and $9.9\times$ over PyTorch and \texttt{scikit-learn}, respectively. The results demonstrate the potential of single-sweep component traversal while identifying on-chip resource constraints as the principal limitation of this approach.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Hardware Architecture (cs.AR); Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2611.02449 — 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 2026-11-03T10:49:38.647Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).