作者: Gennady Livitz , Massimiliano Versace , Anatoly Gorshechnikov , Heather Ames Versace
DOI:
关键词: Set (abstract data type) 、 Massively parallel 、 Cloud computing 、 Computational science 、 Computer science 、 Models of neural computation 、 Architecture 、 Distributed computing 、 Differential equation 、 Dynamical system (definition) 、 Numerical integration
摘要: The system and methods disclosed herein include a runtime architecture that takes nonspecific set of systems differential equations, distributes them across the network, iteratively integrates through time with possibility to output results on every iteration. Embodiments may be used for neural computation or any other suitable application. can as standalone engine part another computational massively parallel numerical integration data-driven dynamical system.