Using the cavity method and diagrammatic methods, we model the dynamics of batch learning of restricted sets of examples, widely applicable to general learning cost functions, and fully taking into account the temporal correlations introduced by the recycling of the examples. The approach is illustrated using the Adaline rule learning teacher-generated or random examples.
The applicability of current seismic-performance-improvement technologies needs to be studied. This research took a super-long-span CFST arch bridge with a total length of 788 m as the object on which to perform a non-linear time-history analysis and a seismic-check calculation according to the seismic response, so as to reveal the seismic weak points of the arch bridge. After the completion of the bridge’s construction, we arranged and utilized the stayed buckle cables (SBCs) reasonably. The seismic performance of the super-long-span CFST arch bridge was improved through friction-pendulum bearings (FPBs) and SBCs. The research shows that FPBs can solve the problem of the insufficient shear resistance of bearings, and SBCs can address the problem whereby the compressive stress of the transverse connection of the main arch exceeds the allowable stress. Moreover, SBCs can increase the transverse stiffness of arch bridges and reduce their seismic responses. Finally, a combination of FPBs and SBCs was adopted to improve the overall seismic performance of the arch bridge and obtain the best seismic-performance-improvement effect.
Appliances energy consumption plays an increasingly important role in the overall building electric energy consumption and its temporal trending. However, predicting appliances energy consumption is complicated by lack of causal understanding of the appliances energy use as well as too many potential predictors that might be relevant to the appliances energy use. In this study, we apply information theory and advance machine learning neural network technique to first rank the importance of potential drivers that dominate appliances energy consumption and secondly model the temporal evolution of appliances energy consumption with a restricted set of environmental predictors. Our results showed that temperature and humidity were the two most important environmental drivers in the house appliances energy consumption modeling. Furthermore, using those environmental drivers, the machine learning model was able to accurately capture the temporal dynamics of appliances energy consumption.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.