arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2002.10214cs.LGcs.AIstat.ML
收藏

用于Transprecision计算的神经网络中的内射领域知识

Injective Domain Knowledge in Neural Networks for Transprecision Computing

University of Bologna(博洛尼亚大学)

Andrea Borghesi, Federico Baldo, Michele Lombardi, Michela Milano

摘要:

机器学习(ML)模型在许多学习任务中非常有效,因为它能够从大数据集中提取有意义的信息。然而,有学习问题,不能很容易地解决依赖于纯数据,例如稀缺的数据或非常复杂的函数近似。幸运的是,在许多情况下,领域知识是明确可用的,可以用来训练更好的ML模型。本文研究了在处理非平凡的学习任务时,通过整合先验知识可以获得的改进,即精度调整的transprecision计算应用程序。域信息以不同的方式注入ML模型中:I)附加特征,II)基于ad-hoc图的网络拓扑,III)正则化方案。结果清楚地表明,利用特定问题信息的ML模型优于纯数据驱动的模型,平均准确率提高了约38%。

英文摘要:

Machine Learning (ML) models are very effective in many learning tasks, due to the capability to extract meaningful information from large data sets. Nevertheless, there are learning problems that cannot be easily solved relying on pure data, e.g. scarce data or very complex functions to be approximated. Fortunately, in many contexts domain knowledge is explicitly available and can be used to train better ML models. This paper studies the improvements that can be obtained by integrating prior knowledge when dealing with a non-trivial learning task, namely precision tuning of transprecision computing applications. The domain information is injected in the ML models in different ways: I) additional features, II) ad-hoc graph-based network topology, III) regularization schemes. The results clearly show that ML models exploiting problem-specific information outperform the purely data-driven ones, with an average accuracy improvement around 38%.

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。

↑