We research the impact of the learning process of neural networks (NN) on the structural properties of the derived graphs. A type of recurrent neural network is used (GARNN). A graph is derived from ...a NN by defining a connection between any pair od nodes having weights in both directions above a certain threshold. We measured structural properties of graphs such as characteristic path lengths (L), clustering coefficients (C) and degree distributions (P). We found that well trained networks differ from badly trained ones in both L and C.
A method for function approximation in reinforcement learning settings is proposed. The action-value function of the Q-learning method is approximated by the radial basis function neural network and ...learned by the gradient descent. Those radial basis units that are unable to fit the local action-value function exactly enough are decomposed into new units with smaller widths. The local temporal-difference error is modelled by a two-class learning vector quantization algorithm, which approximates distributions of the positive and of the negative error and provides the centers of the new units. This method is especially convenient in cases of smooth value functions with large local variation in certain parts of the state space, such that non-uniform placement of basis functions is required. In comparison with four related methods, it has the smallest requirements of basis functions when achieving a comparable accuracy.
A new approach to dynamic systems modeling is given. Stochastic Cellular Automata (SCA) are used as the basic computational module. The dynamic systems are considered as time and space dependent, ...where time dependencies are supposed to be given with some differential equations (DE), while space influences are not known. The basic idea of our approach is to use heuristics for the design of SCA and some stochastic search algorithm to optimize free model parameters. Two non-gradient optimization algorithms are used and evaluated on the two case studies: diffusion and migration of Cs in soil and forest fire spread problem. They are Evolutionary Algorithm (EA) and Stochastic Correlation Algorithm (ALOPEX). We show that with some modifications, both algorithms are capable to solve the two case problems, though there are some important differences between them.
Constructing a single text classifier that excels in any given application is a rather inviable goal. As a result, ensemble systems are becoming an important resource, since they permit the use of ...simpler classifiers and the integration of different knowledge in the learning process. However, many text-classification ensemble approaches have an extremely high computational burden, which poses limitations in applications in real environments. Moreover, state-of-the-art kernel-based classifiers, such as support vector machines and relevance vector machines, demand large resources when applied to large databases. Therefore, we propose the use of a new systematic distributed ensemble framework to tackle these challenges, based on a generic deployment strategy in a cluster distributed environment. We employ a combination of both task and data decomposition of the text-classification system, based on partitioning, communication, agglomeration, and mapping to define and optimize a graph of dependent tasks. Additionally, the framework includes an ensemble system where we exploit diverse patterns of errors and gain from the synergies between the ensemble classifiers. The ensemble data partitioning strategy used is shown to improve the performance of baseline state-of-the-art kernel-based machines. The experimental results show that the performance of the proposed framework outperforms standard methods both in speed and classification.
Novel properties of nanoparticles have numerous potential technological applications but at the same time they underlie new kinds of biological effects. Uniqueness of nanoparticles and nanomaterials ...requires a new experimental methodology. Much evidence suggests that nanoparticles affect cell membrane stability and subsequently exert toxic effects. For this kind of research, lipid vesicles are of high value due to controllability and repeatability of experimental conditions. The aim of work presented here was to develop a computer aided analysis of lipid vesicles shape transformations. We studied a population of palmitoyloleoylphosphatidylcholine (POPC) lipid vesicles after exposure to nanoparticles (C 60 ) or a reference chemical (ZnCl 2 ). With the use of computer image analysis methods, we detected differences in size distributions of vesicles in different exposure groups. Though, at the present state, we are not able to precisely identify effects of nanoparticles on shape transformations of vesicles, those incubated with nanoparticles were in average larger than those in other groups. This population based approach holds many promises for future investigation of nanoparticles-lipid vesicles, or even nanoparticles-biological membranes interactions. However, in order to get reliable results, numerous images have to be analyzed which requires improved and highly automated image segmentation and analyses methods.
The most common type of noise in continuous systems of the real world is Gaussian noise, whereas discrete environments are usually subject to noise of a discrete type. The established, original ...solution for on-line inference of finite automata that is based on generalized recurrent neural networks is evaluated in the presence of noise of both types. It showed quite good performance and robustness.
A new approach is outlined for the digital realization of Boolean neural networks. It is based on programmable gate array technology (PGA). Each cell in the gate array performs the binary ...addition/subtraction function. Space iteration of such a digital circuit allows the calculation of all output functions of the neurons in the network. The topology of the network influences the connections between neurons and therefore the connections in the PGA. This approach offers definite advantages over other solutions.< >