Dombi József (Professzor emeritus, SZTE): Approximation, neural networks, clustering, continuous valued logic

Among the tools of artificial intelligence we find mostly heuristic procedures. Their goodness is difficult to control and a global optimum is not guaranteed. The connection of different algorithms is not solved either. A continuous logic provides a common ground for different algorithms. With the help of it we can also give semantic meaning to the solution; clustering, neural network, approximation, etc. can be handled as well. During the lecture I would like to draw attention to some possible solutions.