We investigate the effect of modular architecture in an artificial neural network for a reinforcement learning problem. Using the supervised backpropagation algorithm to solve a two-task problem, the network performance can be increased by using networks with modular structures. However, using a modular architecture to solve a two-task reinforcement learning problem will not increase the perfor...

The topology and the capacity of a traditional multilayer neural system, as measured by the number of connections in the network, has suprisingly little impact on its generalization ability. This paper presents a new adaptive modular network that offers superior generalization capability. The new network provides significant fault tolerance, quick adaption to novel inputs, and high recognition ...

In a mobile network environment, a Mobile Router (MR) has multiple communication interfaces to the Internet. It switches between or simultaneously uses these interfaces, causing network conditions within the MR to change. If the network conditions were shared with between nodes behind the MR (MNN) and their correspondent nodes (CN), it would be possible for applications to change its behavior a...

This study compares the maze learning performance of three artificial neural network architectures: an Elman recurrent neural network, a long short-term memory (LSTM) network, and Mona, a goal-seeking neural network. The mazes are networks of distinctly marked rooms randomly interconnected by doors that open probabilistically. The mazes are used to examine two important problems related to arti...

In this paper we present several multiple model combination methods, utilizing neural as well as linear predictors, to predict sugar beet crop yield. The results are superior to previous prediction methods which used only neural network or only linear regresison predictors. Abstract. Key Words.

F rom an evolutionary perspective, facial expressions of fear convey highly recognizable emotional signals that serve adaptive functions by promoting survival and reproductive success (1). Current theories of how the brain interprets facial expressions of fear implicate the mirror neuron network (MNN) in echoing the emotional states of others by internal simulation (2,3). Originally discovered ...