Shengquan Ma

Learn More
The nite discrete-time recurrent neural networks are also exploited for potentially innnite computations (e.g. nite automata) where the input is being gradually presented from an external environment via input neurons. Because of gradient learning heuristics or analog hardware implementation reasons the usage of some continuous activation function is(More)
Recently, bipolar fuzzy sets have been studied and applied a bit enthusiastically and a bit increasingly. In this paper we prove that bipolar fuzzy sets and [0,1](2)-sets (which have been deeply studied) are actually cryptomorphic mathematical notions. Since researches or modelings on real world problems often involve multi-agent, multi-attribute,(More)
We consider a multi-layered neural network trained with back propagation. This neural network model realizes a vector function. Sometimes we know the shape of a required function in some area (e.g. known explicit rules in expert system applications) and we want to force it on the neural network during the learning process. We formulate this problem(More)
Let F*(K) be the set of all fuzzy complex numbers. In this paper some classical and measure-theoretical notions are extended to the case of complex fuzzy sets. They are fuzzy complex number-valued distance on F*(K), fuzzy complex number-valued measure on F*(K), and some related notions, such as null-additivity, pseudo-null-additivity, null-subtraction,(More)