In this paper, a new modeling framework of neural network for nonlinear system is proposed. We point out problems in modeling systems with traditional neural networks, that is, difficulty for analyzing internal repres...
详细信息
In this paper, a new modeling framework of neural network for nonlinear system is proposed. We point out problems in modeling systems with traditional neural networks, that is, difficulty for analyzing internal representation, no reproducibility in system modeling (approximation), and no assumption about system property. Based on these considerations, we suggest three main improvements. the first is design of a nonlinear output function. the second is a deterministic scheme for weight initialization. the third is an updating rule for weight parameter. Simulation results show beneficial characteristics of our proposed method.
software reuse includes low-level components reuse, high-level components reuse and system architecture reuse. High-level components reuse and software architecture reuse are still limited to some domain specific mode...
详细信息
ISBN:
(纸本)3540000291
software reuse includes low-level components reuse, high-level components reuse and system architecture reuse. High-level components reuse and software architecture reuse are still limited to some domain specific models, while low-level components reuse is constrained by machine's retrieve ability. this paper proposes a mechanism that builds software in three levels, namely software system, high-level components and low-level components. Each level has a unique structure and organization manner. the focus of the paper is on the construction of high-level components and their matching and composition approaches. Design pattern is proposed for building generic high-level components with large number of alternatives. Once a pattern model of high-level component is constructed, it can be directly used or generalized. Design space incorporated with formal specification technology is introduced to not only precisely describe the relationship between high-level components but also easily analyze components matching and composing. the method is illustrated with a debugger example.
We draw inspiration from properties of "mirror" neurons discovered in the macaque monkey brain area F5, to design and implement a distributed behaviour-based architecture that equips robots with movement imi...
详细信息
We draw inspiration from properties of "mirror" neurons discovered in the macaque monkey brain area F5, to design and implement a distributed behaviour-based architecture that equips robots with movement imitation abilities. We combine this generative route with a learning route, and demonstrate how new composite behaviours that exhibit mirror neuron like properties can be learned from demonstration.
We investigate the use of SVD based two factor models for numerical data classification. Motivations for such a study include the widespread success of such models (e.g, LSI) in textual information retrieval, emerging...
详细信息
We investigate the use of SVD based two factor models for numerical data classification. Motivations for such a study include the widespread success of such models (e.g, LSI) in textual information retrieval, emerging connections with well established statistical techniques and the increasing occurrence of mixed mode (text-and-numeric) data. A direct extension as well as an efficient modification of the LSI model applied to numerical data problems are presented and the associated problems and likely remedies discussed. the techniques under investigation are shown to perform competitively with respect to popular existing numerical classification techniques on a range of synthetic and real world benchmark data. In particular, we show that the modified LSI proposed in this work avoids confronting the optimal subspace selection problem yet generalizes well and remains computationally efficient for large data.
the goal of quality models is to predict a quality factor starting from a set of direct measures. Selecting an appropriate quality model for a particular software is a difficult, non-trivial decision. In this paper, w...
详细信息
the goal of quality models is to predict a quality factor starting from a set of direct measures. Selecting an appropriate quality model for a particular software is a difficult, non-trivial decision. In this paper, we propose an approach to combine and/or adapt existing models (experts) in such way that the combined/adapted model works well on the particular system. Test results indicate that the models perform significantly better than individual experts in the pool.
We present a functional model of V4 area in visual cortex based on predictive coding scheme, in which the prediction is compared withthree kinds of images corresponding to three kinds of image representations which a...
详细信息
We present a functional model of V4 area in visual cortex based on predictive coding scheme, in which the prediction is compared withthree kinds of images corresponding to three kinds of image representations which are projected through the filters withthree different sizes of spatial resolutions. the prediction is represented as a combination of elemental figures. these representations are generated by the response property of main neurons in V1 and V4 areas that the main neurons respond selectively to a limited band of frequencies of spatial brightness distribution. We propose the functional role of elemental figures in invariant perception of object form in visual cortex.
the intrinsic properties of 9;synfire chain9;, the feedforward network propagating synchronous spike packets, has been studied so far. Possible functional roles of the synfire chain, however, has been poorly und...
详细信息
the intrinsic properties of 'synfire chain', the feedforward network propagating synchronous spike packets, has been studied so far. Possible functional roles of the synfire chain, however, has been poorly understood. Considering that coordinated activities of multiple synfire chains can serve as a reference time, we study whether a network model based on the multiple synfire chains contributes to generation of predictive synchrony to occurrence times of external events, observed in the primary motor cortex. In our model, neurons that code occurrence times of external events are partly innervated by the multiple synfire chains. the event times are embedded into the synaptic projections between layers that coincide withthe events and event coding neurons through spike-timing-dependent synaptic learning. From our simulation results, it is found that our model can generate the predictive synchrony when the ratio of the projections is within a suitable range.
An abstract model of a cortical hypercolumn is presented. this model could replicate experimental findings relating to the orientation tuning mechanism in the primary visual cortex. Properties of the orientation selec...
详细信息
An abstract model of a cortical hypercolumn is presented. this model could replicate experimental findings relating to the orientation tuning mechanism in the primary visual cortex. Properties of the orientation selective cells in the primary visual cortex, like contrast-invariance and response saturation, were demonstrated in simulations. We hypothesize that broadly tuned inhibition and local excitatory connections are sufficient for achieving this behavior. We have shown that the local intracortical connectivity of the model is to some extent biologically plausible.
Planning and allocating resources for testing is difficult and it is usually done on an empirical basis, often leading to unsatisfactory results. the possibility of early estimation of the potential faultiness of soft...
详细信息
Planning and allocating resources for testing is difficult and it is usually done on an empirical basis, often leading to unsatisfactory results. the possibility of early estimation of the potential faultiness of software could be of great help for planning and executing testing activities. Most research concentrates on the study of different techniques for computing multivariate models and evaluating their statistical validity, but we still lack experimental data about the validity of such models across different software applications. the paper reports on an empirical study of the validity of multivariate models for predicting software fault-proneness across different applications. It shows that suitably selected multivariate models can predict fault-proneness of modules of different software packages.
this paper introduces a new method for applying multilayer perceptron (MLP) network to control of nonlinear systems. the MLP network is not used directly as a nonlinear controller, but used indirectly via an ARX-like ...
详细信息
ISBN:
(纸本)9810475241
this paper introduces a new method for applying multilayer perceptron (MLP) network to control of nonlinear systems. the MLP network is not used directly as a nonlinear controller, but used indirectly via an ARX-like macro-model. the ARX-like model incorporating MLP network is constructed in such a way that it has similar linear properties to a linear ARX model. the nonlinear controller is then designed in the same way as designing a linear controller based on a linear ARX model. Numerical simulations are carried to demonstrate the effectiveness of the new method.
暂无评论