Comments on “Can backpropagation error surface not have localminima?”

in the above paper Yu (IEEE Trans. Neural Networks, vol.3, no.6, p.1019-21 (1992)) claims to prove that local minima do not exist in the error surface of backpropagation networks being trained on data with t distinct input patterns when the network is capable of exactly representing arbitrary mappings on t input patterns. The commenter points out that the proof presented is flawed, so that the resulting claims remain unproved. In reply, Yu points out that the undesired phenomenon that was sited can be avoided by simply imposing the arbitrary mapping capacity of the network on lemma 1 in the article.     

Hierarchical models for analysis and recognition of handwritten characters

Different hierarchical models in pattern analysis and recognition are proposed, based on occurrence probability of patterns. As an important application of recognizing handprinted characters, three typical kinds of hierarchical models such asM _89-89,M _89-36 andM _36-36 have been presented, accompanied by the computer algorithms for computing recognition rates of pattern parts. Moreover, a comparative study of their recognition rates has been conducted theoretically; and numerical experiments have been carried out to verify the analytical conclusions made. Various hierarchical models deliberated in this paper can provide users more or better choices of pattern models in practical application, and lead to a uniform computational scheme (or code). The recognition rates of parts can be improved remarkably by a suitable hierarchical model. For the modelM _89-36 in which case some of the Canadian standard handprinted characters have multiple occurrence probabilities, the total mean recognition rates of the given sample may reach 120% of that by the model proposed by Li et al., and 156% of that obtained from the subjective experiments reported by Suen.     

First break refraction event picking using fuzzy logic systems

First break picking is a pattern recognition problem in seismic signal processing, one that requires much human effort and is difficult to automate. The authors' goal is to reduce the manual effort in the picking process and accurately perform the picking. Feedforward neural network first break pickers have been developed using backpropagation training algorithms applied either to an encoded version of the raw data or to derived seismic attributes which are extracted from the raw data. The authors summarize a study in which they applied a backpropagation fuzzy logic system (BPFLS) to first break picking. The authors use derived seismic attributes as features, and take lateral variations into account by using the distance to a piecewise linear guiding function as a new feature. Experimental results indicate that the BPFLS achieves about the same picking accuracy as a feedforward neural network that is also trained using a backpropagation algorithm; however, the BPFLS is trained in a much shorter time, because there is a systematic way in which the initial parameters of the BPFLS can be chosen, versus the random way in which the weights of the neural network are chosen     

A coverage plan for health centres in Murewa district in Zimbabwe: an example of action research

Good access to health facilities providing good first-level health care remains problematic in many developing countries. It is a hindrance to effective and efficient functioning of the hospital, as outpatient departments become overcrowded with patients from areas without health centres. In many cases the quality of care delivered to these patients is poor because within the district health system the hospital is not the best place for the supply of comprehensive, integrated and continuous care. Eventually, high hospital involvement in first-level care can jeopardize the delivery of adequate referral care for those patients who desperately need the hospital's technology and expertise. This paper provides an account of the way this problem was investigated and managed by the district health management team in the Murewa district in north-east Zimbabwe. The design of a comprehensive 'master plan' or 'coverage plan' is presented as well as the problems and difficulties encountered. The Murewa experience highlights the relevance of a coverage plan for rational and coherent health infrastructure planning at district level. The approach followed by the Murewa team illustrates the use of action research as an integral part of the management of district health systems.     

Mechanism for self-synchronization of femtosecond pulses in atwo-color Ti:sapphire laser

An experimental and theoretical analysis of the nonlinear coupling mechanism between the two solitary pulses circulating in a two-color femtosecond laser is presented. Two operation regimes; synchronized; and nonsynchronized; and a hysteresis of the transition between the two regimes are clearly observed; while independent modelocking and tunability of the output pulse trains is found in both regimes. Pulses in the range from 15 to 100 fs are synchronized with a timing jitter below 2 fs. The combined effects of cross-phase modulation and negative group velocity dispersion are shown to be responsible for the strong pulse correlation in the synchronized regime. Our experimental observations are in agreement with numerical simulations, thus confirming the theoretical model     

Modeling of Diffusion in Capillary Porous Materials During the Drying Process

This paper presents a model of heterogenous diffusion in capillary porous materials during the process of drying. The governing heat and mass transfer equations have been established using the liquid as well as vapor flow. Two models have been presented. Model 1 does not consider the heat conduction while the model 2 has been established by considering the conduction. The developed models and the numerical solutions of the resulting differential equations can take into account the moisture and temperature dependent thermophysical properties of the product. All equations have been established in spherical coordinates but the programme written for the purpose of calculations can be used for other geometries also. Numerical calculations have been performed for gas concrete and tiles using model 1, while model 2 has been used for gas concrete only because of the lack of data for thermophysical properties of the tile. For gas concrete it was seen that conduction has only marginal effect on the drying process and the numerical predictions of the drying process were reasonably accurate.