Search

Abstract: Reservoir computing is a framework for computation like a recurrent neural network that allows for the black box modeling of dynamical systems. In contrast to other recurrent neural network approaches, reservoir computing does not train the input and internal weights of the network, only the readout is trained. However it is necessary to adjust parameters to create a “good” reservoir for a given application. In this study we introduce a method, called RCDESIGN (reservoir computing and design training). RCDESIGN combines an evolutionary algorithm with reservoir computing and simultaneously looks for the best values of parameters, topology and weight matrices without rescaling the reservoir matrix by the spectral radius. The idea of adjust the spectral radius within the unit circle in the complex plane comes from the linear system theory. However, this argument does not necessarily apply to nonlinear systems, which is the case of reservoir computing. The results obtained with the proposed method are compared with results obtained by a genetic algorithm search for global parameters generation of reservoir computing. Four time series were used to validate RCDESIGN. [Copyright &y& Elsevier]

Abstract Heat waves give rise to order of magnitude higher mortality rates than other weather-related natural disasters. Unfortunately both the severity and amplitude of heat waves are predicted to increase worldwide as a consequence of climate change. Hence, meteorological services have a growing need to identify such periods in order to set alerts, whilst researchers and industry need representative future heat waves to study risk. This paper introduces a new location-specific mortality risk focused definition of heat waves and a new mathematical framework for the creation of time series that represents them. It focuses on identifying periods when temperatures are high during the day and night, as this coincidence is strongly linked to mortality. The approach is tested using observed data from Brazil and the UK. Comparisons with previous methods demonstrate that this new approach represents a major advance that can be adopted worldwide by governments, researchers and industry. Highlight • A novel weather file for building assessment facing high temperature scenarios is proposed. • The method relies on KDD involving quantile regression ensemble and data integration. • It provides additional knowledge of inputs that affect temperature extremes. • The outcome is validated w.r.t. standards and is ready to use by industry. [ABSTRACT FROM AUTHOR]