Short-term Cognitive Networks, Flexible Reasoning and Nonsynaptic Learning

While the machine learning literature dedicated to fully automated reasoning algorithms is abundant, the number of methods enabling the inference process on the basis of previously defined knowledge structures is scanter. Fuzzy Cognitive Maps (FCMs) are recurrent neural networks that can be exploited towards this goal because of their flexibility to handle external knowledge. However, FCMs suffer from a number of issues that range from the limited prediction horizon to the absence of theoretically sound learning algorithms able to produce accurate predictions. In this paper we propose a neural system named Short-term Cognitive Networks that tackle some of these limitations. In our model, used for regression and pattern completion, weights are not constricted and may have a causal nature or not. As a second contribution, we present a nonsynaptic learning algorithm to improve the network performance without modifying the previously defined weight matrix. Besides, we derive a stop condition to prevent the algorithm from iterating without significantly decreasing the global simulation error.