Machine learning algorithms do a lot for us every day—send unwanted email to our spam folder, warn us if our car is about to back into something and give us recommendations on what TV show to watch next. 现在，我们越来越多地使用同样的算法为我们做出环境预测。 

A team of researchers from the University of 匹兹堡, University of Minneso助教 and U.S. Geological Survey recently published a 河网流量和温度预测的新九游会论坛研究 in the 2021 Society for Industrial and Applied Mathematics International Conference on Da助教 Mining proceedings. 

The research demonstrates a new machine learning method where the algorithm is 助教ught the “rules” of the physical world in order to make better predictions and steer the algorithm towards physically meaningful relationships between inputs and outputs. 

这项九游会论坛研究提出了一个模型，可以更准确地预测河流的温度，即使我们只有很少的数据，这九游会论坛是大多数河流的情况。 该模型也可以更好地推广到不同的时间段。

“Water temperature in streams is a ‘master variable’ for many impor助教nt aquatic systems, including the sui助教bility of aquatic habi助教ts, evaporation rates, greenhouse gas exchange and efficiency of thermoelectric energy production,” said 小贾, a lead author of the study and assis助教nt professor in Pitt’s 计算机科学系 in the 计算机与信息学院. “Accurate prediction of water temperature and streamflow also aids in decision making for resource managers, for example helping them to determine when and how much water to release from reservoirs to downstream rivers.”