利用语音学知识和神经网络识别非特定人的复合元音

本文主要研究：以音素为识别基元，运用语音学知识，对非特定人的普通话复合元音进行识别。其特点是音素识别由神经网络（ＮＮ）完成，为了便于利用语音知识ＮＮ输入层的刺激采用语音的功率谱，用单元音训练的ＮＮ识另非特定人的普通话复合元音，识别率是５４％。而运用语音学知识后，其识别率提高到９０％。     

时延细胞神经网络的概周期解的存在性和指数稳定性

研究时延细胞神经网络的概周期解存在性和全局指数稳定性问题 ,巧妙地引入可调实参数di>0 (i =1,2 ,… ,n) ,获得了该神经网络存在唯一的概周期解的充分条件和所有其它解均指数地收敛于此概周期解的充分条件 .     

Multiple spatial coherence resonance induced by the stochastic signal in neuronal networks near a saddle-node bifurcation

This paper reported multiple induction of spiral waves with a stochastic signal in a square lattice network model composed of type I Morris-Lecar (ML) neurons, where each neuron is coupled to its four nearest neighbors. The induction occurs in two or three distinct regions of noise intensity, and thus enables emergence of multiple spatial coherence in the network, demonstrating a novel evidence of multiple coherence resonance. Emergence of this multiple spatial coherence resonance was evidenced by calculating the degree of spatial complexity, spatial correlation length, spatial structure function, circular symmetry, and signal-to-noise ratio curves. The network was further characterized by spatial frequency and inherent spatial scale, reflecting its inherent ability to manifest ordered pattern formation under the driven of noisy signals.     

First principles versus artificial neural network modelling of a solar desalination system with experimental validation

ABSTRACT

The present study mainly focuses on enhancing the performance of solar still unit using solar energy through cylindrical parabolic collector and solar panels. A 300 W solar panel is used to heat saline water by thermal elements outside the solar still unit. Solar panels are cooled during the hot hours of the day; thus, reducing their temperature may lead to an increase in solar panel efficiency followed by an increase in the efficiency of the solar still unit. The maximum amount of freshwater used in the experiment was 2.132 kg/day. The experiments were modelled using ANNs. Based on neural network simulation results, there is a significant correlation between experimental data and neural network modelling. This paper compares experimental data with data obtained from mathematical modelling and ANNs. As a conclusion, the artificial neural network prediction has been more accurate than the simplified first principles model presented.     

Coexistence of domains of attraction of nonautonomous neural networks with time-varying delays

This paper presents new dynamical behavior, i.e., the coexistence of 2^N domains of attraction of N-dimensional nonautonomous neural networks with time-varying delays. By imposing some new assumptions on activation functions and system parameters, we construct 2^N invariant basins for neural system and derive some criteria on the boundedness and exponential attractivity for each invariant basin. Particularly, when neural system degenerates into periodic case, we not only attain the coexistence of 2^N periodic orbits in bounded invariant basins but also give their domains of attraction. Moreover, our results are suitable for autonomous neural systems. Our new results improve and generalize former ones. Finally, computer simulation is performed to illustrate the feasibility of our results.     

An approximation-based approach for fuzzy multi-period production planning problem with credibility objective

This paper develops a fuzzy multi-period production planning and sourcing problem with credibility objective, in which a manufacturer has a number of plants or subcontractors. According to the credibility service levels set by customers in advance, the manufacturer has to satisfy different product demands. In the proposed production problem, production cost, inventory cost and product demands are uncertain and characterized by fuzzy variables. The problem is to determine when and how many products are manufactured so as to maximize the credibility of the fuzzy costs not exceeding a given allowable invested capital, and this credibility can be regarded as the investment risk criteria in fuzzy decision systems. In the case when the fuzzy parameters are mutually independent gamma distributions, we can turn the service level constraints into their equivalent deterministic forms. However, in this situation the exact analytical expression for the credibility objective is unavailable, thus conventional optimization algorithms cannot be used to solve our production planning problems. To overcome this obstacle, we adopt an approximation scheme to compute the credibility objective, and deal with the convergence about the computational method. Furthermore, we develop two heuristic solution methods. The first is a combination of the approximation method and a particle swarm optimization (PSO) algorithm, and the second is a hybrid algorithm by integrating the approximation method, a neural network (NN), and the PSO algorithm. Finally, we consider one 6-product source, 6-period production planning problem, and compare the effectiveness of two algorithms via numerical experiments.     

Data Envelopment Analysis of multiple response experiments

Taguchi method is the usual strategy in robust design and involves conducting experiments using orthogonal arrays and estimating the combination of factor levels that optimizes a given performance measure, typically a signal-to-noise ratio. The problem is more complex in the case of multiple responses since the combinations of factor levels that optimize the different responses usually differ. In this paper, an Artificial Neural Network, trained with the experiments results, is used to estimate the responses for all factor level combinations. After that, Data Envelopment Analysis (DEA) is used first to select the efficient (i.e. non-dominated) factor level combinations and then for choosing among them the one which leads to a most robust quality loss penalization. Mean Square Deviations of the quality characteristics are used as DEA inputs. Among the advantages of the proposed approach over traditional Taguchi method are the non-parametric, non-linear way of estimating quality loss measures for unobserved factor combinations and the non-parametric character of the performance evaluation of all the factor combinations. The proposed approach is applied to a number of case studies from the literature and compared with existing approaches.     

Reliable computing with unreliable components: Using separable environments to stabilize long-term information storage

How, in the face of both intrinsic and extrinsic volatility, can unconventional computing fabrics store information over arbitrarily long periods? Here, we argue that the predictable structure of many realistic environments, both natural and artificial, can be used to maintain useful categorical boundaries even when the computational fabric itself is inherently volatile and the inputs and outputs are partially stochastic. As a concrete example, we consider the storage of binary classifications in connectionist networks, although the underlying principles should be applicable to other unconventional computing paradigms. Specifically, we demonstrate that an unsupervised, activity dependent plasticity rule, AHAH (Anti-Hebbian-And-Hebbian), allows binary classifications to remain stable even when the underlying synaptic weights are subject to random noise. When embedded in environments composed of separable features, the weight vector is restricted by the AHAH rule to local attractors representing stable partitions of the input space, allowing unsupervised recovery of stored binary classifications following random perturbations that leave the system in the same basin of attraction. We conclude that the stability of long-term memories may depend not so much on the reliability of the underlying substrate, but rather on the reproducible structure of the environment itself, suggesting a new paradigm for reliable computing with unreliable components.