摘要
由一名新闻记者-机器人与机器学习的工作人员新闻编辑每日新闻-机器学习的新数据在一份新的报告中提供。根据NewsRx记者从湖北发回的新闻报道，研究表明：“土壤-植物-大气连续体(SPAC)系统的传统模型是以物理为基础的，其实际应用受到参数化程度高、结构偏差大、运行成本高等问题的阻碍。”"针对传统建模的困难,提出了一种基于高斯过程(GP)的机器学习建模方法."本研究的资助者包括宁夏重点研究开发项目，国家自然科学基金(NSFC)。本报编辑引用武汉大学的一篇研究文章:“同时,指出了传统机器学习在可互斥性和机制表示方面的不足,提出了构造局部GP组框架和改进每个局部GP中核函数的创新方法,该模型具有较高的能力和较高的鲁棒性。”包括表示复杂SPAC S系统内子进程之间的相互作用，表示由作物生长进度变化引起的非平稳子进程动态，在包含三种不同SOI L条件的合成SPAC系统数据集上测试了该模型的性能，结果表明，该模型对子过程机制的解释在不同土壤条件下是稳健的，与传统的全局GP模型和两种深度学习模型(DNN)相比较，该模型对子过程机制的解释是一致的。我们的模型不仅在常规预测实验中表现得更好，而且在需要模型在不同条件下传递的泛化实验中表现得更好。我们的研究表明，通过改进特征表示函数，IT有望使机器学习模型具有可解释性。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Fresh data on Machine Learning are pre sented in a new report. According to news reporting originating from Hubei, Peop le's Republic of China, by NewsRx correspondents, research stated, "Traditional models of the soil -plant -atmosphere continuum (SPAC) system are physics -based and their practical application has been hindered by issues of high parameteriz ation, prior structural bias, and costly running. In this paper, we developed a machine learning modelling method based on the Gaussian process (GP) to avoid th ese difficulties of traditional modelling." Funders for this research include Priority Research and Development Projects for Ningxia, National Natural Science Foundation of China (NSFC). Our news editors obtained a quote from the research from Wuhan University, "At t he same time, shortcomings of conventional machine learning in terms of interpre tability and mechanism representation were addressed. The innovative method cons isted of structuring a local GP group framework and improving kernel functions u sed in each local GP. The resultant model was endowed with advanced capacities, including representing interplays between subprocesses within the complex SPAC s ystem, representing nonstationary subprocess dynamics caused by crop growth stag e shifts, as well as automatically interpreting key low -order input interaction s and dominant input variables for each subprocess. The performance of our model was examined on synthetic SPAC system datasets that covered three different soi l conditions. Results demonstrated that its interpretations regarding subprocess mechanisms were robust across different soil conditions and consistent with dom ain knowledge. Compared to a conventional global GP model and two deep learning models (DNN and LSTM), our model performed significantly better not only in regu lar prediction experiments but also in generalization experiments that required models to be transferred across different conditions. Our study suggested that i t is promising to enable a machine learning model interpretable by improving its feature representation function."
NETL
NSTL