arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.16340cs.LGmath.OCstat.ML
收藏

经典机器学习问题在GPU上的优化

Optimization for Classical Machine Learning Problems on the GPU

Friedrich-Schiller-University Jena(耶拿弗里德里希·席勒大学);Data Assessment Solutions GmbH(数据评估解决方案有限责任公司)

Sören Laue, Mark Blacher, Joachim Giesen

摘要:

约束优化问题经常出现在经典的机器学习中。存在解决约束优化的框架,例如CVXPY和GENO。然而,与深度学习框架相比,GPU支持有限。在这里,我们扩展了GENO框架,也可以在GPU上解决约束优化问题。该框架允许用户以易于阅读的建模语言指定约束优化问题。然后,求解器将根据此规范自动生成。当在GPU上运行时,该求解器的性能优于CVXPY等最先进的方法与cuOSQP或SCS等GPU加速求解器的组合。

英文摘要:

Constrained optimization problems arise frequently in classical machine learning. There exist frameworks addressing constrained optimization, for instance, CVXPY and GENO. However, in contrast to deep learning frameworks, GPU support is limited. Here, we extend the GENO framework to also solve constrained optimization problems on the GPU. The framework allows the user to specify constrained optimization problems in an easy-to-read modeling language. A solver is then automatically generated from this specification. When run on the GPU, the solver outperforms state-of-the-art approaches like CVXPY combined with a GPU-accelerated solver such as cuOSQP or SCS by a few orders of magnitude.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑