A neural network algorithm for the multiple traveling salesmen problem

We developed an efficient neural network algorithm for solving the Multiple Traveling Salesmen Problem (MTSP). A new transformation of the N-city M-salesmen MTSP to the standard Traveling Salesmen Problem (TSP) is introduced. The transformed problem is represented by an expanded version of Hopfield-Tank's neuromorphic city-position map with (N + M-1)-cities and a single fictitious salesmen. The dynamic model associated with the problem is based on the Basic Differential Multiplier Method (BDMM) [26] which evaluates Lagrange multipliers simultaneously with the problem's state variables. The algorithm was successfully tested on many problems with up to 30 cities and five salesmen. In all test cases, the algorithm always converged to valid solutions. The great advantage of this kind of algorithm is that it can provide solutions to complex decision making problems directly by solving a system of ordinary differential equations. No learning steps, logical if statements or adjusting of parameters are required during the computation. The algorithm can therefore be implemented in hardware to solve complex constraint satisfaction problems such as the MTSP at the speed of analog silicon VLSI devices or possibly future optical neural computers.     

On problem solving with Hopfield neural networks

Hopfield and Tank have shown that neural networks can be used to solve certain computationally hard problems, in particular they studied the Traveling Salesman Problem (TSP). Based on network simulation results they conclude that analog VLSI neural nets can be promising in solving these problems. Recently, Wilson and Pawley presented the results of their simulations which contradict the original results and cast doubts on the usefulness of neural nets. In this paper we give the results of our simulations that clarify some of the discrepancies. We also investigate the scaling of TSP solutions found by neural nets as the size of the problem increases. Further, we consider the neural net solution of the Clustering Problem, also a computationally hard problem, and discuss the types of problems that appear to be well suited for a neural net approach.     

Clustering with neural networks

Partitioning a set ofN patterns in ad-dimensional metric space intoK clusters — in a way that those in a given cluster are more similar to each other than the rest — is a problem of interest in many fields, such as, image analysis, taxonomy, astrophysics, etc. As there are approximatelyK ^N/K! possible ways of partitioning the patterns amongK clusters, finding the best solution is beyond exhaustive search whenN is large. We show that this problem, in spite of its exponential complexity, can be formulated as an optimization problem for which very good, but not necessarily optimal, solutions can be found by using a Hopfield model of neural networks. To obtain a very good solution, the network must start from many randomly selected initial states. The network is simulated on the MPP, a 128 × 128 SIMD array machine, where we use the massive parallelism not only in solving the differential equations that govern the evolution of the network, but also in starting the network from many initial states at once thus obtaining many solutions in one run. We achieve speedups of two to three orders of magnitude over serial implementations and the promise through Analog VLSI implementations of further speedups of three to six orders of magnitude.Supported by a National Research Council-NASA Research Associatship     

Radial basis function neural networks for modeling growth rates of the basidiomycetes Physisporinus vitreus and Neolentinus lepideus

A radial basis function (RBF) neural network was developed and compared against a quadratic response surface (RS) model for predicting the specific growth rates of the biotechnologically important basidiomycetous fungi, Physisporinus vitreus and Neolentinus lepideus, under three environmental conditions: temperature (10–30 °C), water activity (0.950–9.998), and pH (4–6). Both the RBF network and polynomial RS model were mathematically evaluated against experimental data using graphical plots and several statistical indices. The evaluation showed that both models gave reasonably good predictions, but the performance of the RBF neural network was superior to that of the classical statistical method for all three data sets used (training, testing, full). Sensitivity analysis revealed that of the three experimental factors the most influential on the growth rate of P. vitreus was water activity, followed by temperature and pH to a lesser extent. In contrast, temperature in particular and then water activity were the key determinants of the development of N. lepideus. RBF neural networks could be a powerful technique for modeling fungal growth behavior under certain parameters and an alternative to time-consuming, traditional microbiological techniques.     

Curcumin increased the expression of c-FLIP in HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP) patients

Human T-cell lymphotropic virus type 1 (HTLV-1)-associated myelopathy/tropical spastic paraparesis (HAM/TSP) disease is a chronic neuroinflammatory disease, which is associated with HTLV-1 infection. There is no effective and satisfactory treatment of HAM/TSP. It has been shown that curcumin exhibits modulatory effects on apoptosis and cytotoxicity-related molecules in HAM/TSP patients. In the present study, we examined the effect of curcumin on the gene expression of caspase-8, caspase-10, and anti-apoptotic protein c-FLIP, in HAM/TSP patients. Furthermore, we compared the expression of these molecules between HAM/TSP and asymptomatic carriers. Real-time PCR was performed to examine the mRNA expression of caspase-8, caspase-10, and c-FLIP in studied groups. The mRNA expression of caspase-8 and caspase-10 was similar before and after curcumin treatment in HAM/TSP patients (P > 0.05). The mRNA expression of c-FLIPL and c-FLIPs was higher after curcumin treatment compared with before treatment and significant differences were observed between the two groups (P = 0.004 and P = 0.044, respectively). The mRNA expression levels of caspase-8, caspase-10, c-FLIPL, and c-FLIPs were not statistically significant between HAM/TSP patients and asymptomatic carriers (P < 0.05). In conclusion, our results showed that curcumin increased the expression of c-FLIP in HAM/TSP patients which might suggest that, this molecule is involved in the apoptosis of HTLV-1-infected cells. Further studies with large sample size could be useful to clarify the role of this supplement in HAM/TSP patients.     

“Turtle watching” conservation guidelines: green turtle (<Emphasis Type="Italic">Chelonia mydas</Emphasis>) tourism in nearshore coastal environments

We propose a conservational opportunity for humans to ‘use’ the green turtle (Chelonia mydas) in a non-consumptive manner. Although the concept of a social safe-minimum standard analysis, as applied to the sustainability of tourism-dependent turtle watching, has focused on beach-nesting habitats, other tourist activities like diving and snorkelling also occur in shallow coastal habitats frequented by juvenile and adult turtles. When integrated over time, at a specific location, such tourism activities may compromise turtle physiology in a manner that limits conservation goals for the species and hence the tourism. We identify research insights that can be used to achieve a creatively managed tourism—one that allows tourists to observe turtles in their natural coastal habitat in a manner that is commensurate with functional turtle conservation. We propose management options loosely based on whale-watching: i.e. voluntary and/or mandatory regulations based on home-range studies that identify localized temporal and spatial patterns of habitat use exhibited by turtles. We recommend temporally- and spatially-dynamic stratified-random-design tours that exclude critical local (small-scale) habitat and include less-critical habitat on a randomized rotational basis. Practical guidelines for tour operators that are founded on turtle habitat-occupancy patterns may ensure expanded life-history conservation measures and sustainable turtle-watching tourism.     

Stability analysis of higher-order neural networks for combinatorial optimization

Recurrent neural networks with higher order connections, from here on referred to as higher-order neural networks (HONNs), may be used for the solution of combinatorial optimization problems. In Ref. 5 a mapping of the traveling salesman problem (TSP) onto a HONN of arbitrary order was developed, thereby creating a family of related networks that can be used to solve the TSP. In this paper, we explore the trade-off between network complexity and quality of solution that is made available by the HONN mapping of the TSP. The trade-off is investigated by undertaking an analysis of the stability of valid solutions to the TSP in a HONN of arbitrary order. The techniques used to perform the stability analysis are not new, but have been widely used elsewhere in the literature. The original contribution in this paper is the application of these techniques to a HONN of arbitrary order used to solve the TSP. The results of the stability analysis show that the quality of solution is improved by increasing the network complexity, as measured by the order of the network. Furthermore, it is shown that the Hopfield network, as the simplest network in the family of higher-order networks, is expected to produce the poorest quality of solution.     

Evolution in a chronic RNA virus infection: selection on HTLV-I tax protein differs between healthy carriers and patients with tropical spastic paraparesis

HTLV-I causes T-cell leukemia and tropical spastic paraparesis (TSP) in a minority of infected people, whereas the majority remain healthy. The virus differs little in sequence between isolates but has been shown to have a quasispecies structure. Using the Nei and Gojobori algorithm, we have shown that the proportion of nonsynonymous to synonymous changes in HTLV-I proviral tax gene sequences from healthy seropositive subjects (Dn/Ds= 0.9 to 1.3) is significantly higher than those from TSP patients (Dn/Ds= 0.3 to 0.6). Here we show that the distinction between healthy seropositives and TSP patients can only be seen with proviral tax sequences, but not with cDNA, the amino-terminal or carboxy-terminal half of tax, or the rex gene. The Dn/Ds ratio of proviral tax sequences was used to analyze two TSP patients with atypical features and to investigate the influence of cytotoxic T cells (CTL) on the viral quasispecies. Received: 18 January 1995 / Accepted: 7 November 1995     

Using a neural network to identify potential HLA-DR1 binding sites within proteins

The presentation by antigen-presenting cells of immunodominant peptide segments in association with major histocompatibility complex (MHC) encoded proteins is fundamental to the efficacy of a specific immune response. One approach used to identify immunodominant segments within proteins has involved the development of predictive algorithms which utilize amino acid sequence data to identify structural characteristics or motifs associated with in vivo antigenicity. The parallel-computing technique termed ‘neural networking’ has recently been shown to be remarkably efficient at addressing the problem of pattern recognition and can be applied to predict protein secondary structure attributes directly from amino acid sequence data. In order to examine the potential of a neural network to generalize peptide structural feature related to binding within class II MHC-encoded proteins, we have trained a neural network to determine whether or not any given amino acid of a protein is part of a peptide segment capable of binding to HLA-DR1. We report that a neural network trained on a data base consisting of peptide segments known to bind to HLA-DR1 is able to generalize features relating to HLA-DR1-binding capacity (r = 0.17 and p = 0.0001).     

高阶Boltzmann机的平均场理论学习算法

用统计力学中的平均场理论及模拟退火技术,将一般高阶神经网络及玻尔兹曼机的优点结合在一起,以不同于文献［２］的方法,推导出高阶玻尔兹曼机的驰豫动力学的确定性方程和平均场理论学习算法。二者皆便于ＶＬＳＩ实现,且学习算法省去很多ＣＰＵ时间,对二维镜像对称问题及Ｔ－问题的计算机仿真结果表明三阶玻尔兹曼机的平均场理论学习算法是正确的,且性能较二阶玻尔兹曼机好。     

Coupling isoprene and monoterpene emissions from Amazonian tree species with physiological and environmental parameters using a neural network approach

The ability to predict isoprene emissions from plants is important for predicting atmospheric chemistry. To improve the basis for prediction capability, data obtained from continuous field measurements of isoprene and monoterpene emissions from three Amazonian tree species were related to observed environmental and leaf physiological parameters using a new neural network approach. The environmental parameters included leaf temperature, light, relative humidity, water vapour pressure deficit, and the history of ambient temperature and ozone concentration, whereas the physiological parameters included stomatal conductance, assimilation and intercellular CO₂ concentration. The neural approach with 24 different combinations of these parameters was applied to predict the emission variability observed during short time periods (2–3 d) with individual tree branches and, on a longer-term scale, in aggregated data sets from different seasons, leaf developmental stage, and light environment. The results were compared to the quasi standard emission algorithm for isoprene. On the short-term scale, good agreement (r²≈ 0.9) was obtained between observations and predictions of the standard algorithm as well as predictions of the neural network using the same input parameters (leaf temperature and light). When these predictors were used to model the long-term emission variability, r² was reduced to < 0.5 for both approaches. Remarkably, for the neural technique, more than 50% of the unexplained variance could be explained by the mean temperature of the preceding 36 h. An even better network performance was obtained with physiological parameter combinations (r² > 0.9) suggesting a strong and applicable link between isoprenoid emission and leaf primary metabolism.     

Temperature coupling in cricket acoustic communication

Summary Temperature effects on calling song production and recognition were investigated in the North American field cricket, Gryllus firmus. Temporal parameters of field-recorded G. firmus calling song are strongly affected by temperature. Chirp rate and syllable rate increase, by factors of 4 and 2, respectively, as linear functions of temperature over the range in which these animals sing in the field (12°–30 °C). Temperature affects syllable duration to a lesser extent, and does not influence calling song carrier frequency. Female phonotactic preference, measured on a spherical treadmill in the laboratory, also changes with temperature such that warmer females prefer songs with faster chirp and syllable rates. Best phonotaxis, measured as accuracy of orientation to the sound source, and highest walking velocity, occur in response to temperature-matched songs at 15°, 21°, and 30 °C. Experiments under semi-natural conditions in an outdoor arena revealed that females perform phonotaxis at temperatures as low as 13 °C. Taken together, the song and phonotaxis data demonstrate that this communication system is temperature coupled. A strategy is outlined by which temperature coupling may be exploited to test hypotheses about the organization of neural networks subserving song recognition.Abbreviations CP chirp period - SP syllable period - SD syllable duration     

Pattern-recognition by an artificial network derived from biologic neuronal systems

A novel artificial neural network, derived from neurobiological observations, is described and examples of its performance are presented. This DYnamically STable Associative Learning (DYSTAL) network associatively learns both correlations and anticorrelations, and can be configured to classify or restore patterns with only a change in the number of output units. DYSTAL exhibits some particularly desirable properties: computational effort scales linearly with the number of connections, i.e., it is0(N) in complexity; performance of the network is stable with respect to network parameters over wide ranges of their values and over the size of the input field; storage of a very large number of patterns is possible; patterns need not be orthogonal; network connections are not restricted to multi-layer feed-forward or any other specific structure; and, for a known set of deterministic input patterns, the network weights can be computed, a priori, in closed form. The network has been associatively trained to perform the XOR function as well as other classification tasks. The network has also been trained to restore patterns obscured by binary or analog noise. Neither global nor local feedback connections are required during learning; hence the network is particularly suitable for hardware (VLSI) implementation.     

Ribosome biogenesis is temperature‐dependent and delayed in Escherichia coli lacking the chaperones DnaK or DnaJ†

In Escherichia coli strains carrying null mutations in either the dnaK or dnaJ genes, the late stages of 30S and 50S ribosomal subunit biogenesis are slowed down in a temperature‐dependent manner. At high temperature (44°C), 32S and 45S particles (precursors to 50S subunits) and 21S particles (precursors to 30S subunits) accumulate. The latter are shown by 3′5′ rapid amplification of cDNA ends analysis to contain unprocessed or partially processed 16S ribosomal RNA at the 5′ end, but the 3′ end was never processed. This implies that maturation of 16S ribosomal RNA starts at the 5′‐terminus, and that the 3′‐terminus is only trimmed at a later step. At normal temperatures (30°C?37°C), ribosome assembly in both mutants is not arrested but is significantly delayed, as shown by pulse‐chase analysis. Assembly defects are partially compensated by an overexpression of other heat‐shock proteins, which occurs in the absence of their negative regulator DnaK, or by a plasmid‐driven overexpression of GroES/GroEL, suggesting the involvement of a network of chaperones in ribosome biogenesis.     

Development of diatom-based tools for assessing stream water quality in south-eastern Australia: assessment of environmental transfer functions

Models were developed of relationships among water quality, geospatial and species-level diatom data for 465 samples collected from stream sites across south-eastern Australia. Transfer functions were derived from weighted averaging and artificial neural network approaches. Analysis of spatial variations in species assemblages was used to divide the sites into two groups according to site elevation. The strongest predictive models for the upland group associated diatom assemblages with conductivity, longitude, altitude, and to a lesser extent pH, NO _x and TKN. The strongest predictors for the lowland group were longitude and conductivity, but artificial neural network models performed well for NO _x and temperature. The importance of the geospatial variables suggests that there may be a capacity to develop diatom sub-regions within which robust models for other water quality variables important to management can be generated. Electronic supplementary material Electronic supplementary material is available for this article at and accessible for authorised users. Handling editor: R. Norriss     

Decreased astrocytic thrombospondin‐1 secretion after chronic ammonia treatment reduces the level of synaptic proteins: in vitro and in vivo studies

Chronic hepatic encephalopathy (CHE) is a major complication in patients with severe liver disease. Elevated blood and brain ammonia levels have been implicated in its pathogenesis, and astrocytes are the principal neural cells involved in this disorder. Since defective synthesis and release of astrocytic factors have been shown to impair synaptic integrity in other neurological conditions, we examined whether thrombospondin‐1 (TSP‐1), an astrocytic factor involved in the maintenance of synaptic integrity, is also altered in CHE. Cultured astrocytes were exposed to ammonia (NH₄Cl, 0.5–2.5 mM) for 1–10 days, and TSP‐1 content was measured in cell extracts and culture media. Astrocytes exposed to ammonia exhibited a reduction in intra‐ and extracellular TSP‐1 levels. Exposure of cultured neurons to conditioned media from ammonia‐treated astrocytes showed a decrease in synaptophysin, PSD95, and synaptotagmin levels. Conditioned media from TSP‐1 over‐expressing astrocytes that were treated with ammonia, when added to cultured neurons, reversed the decline in synaptic proteins. Recombinant TSP‐1 similarly reversed the decrease in synaptic proteins. Metformin, an agent known to increase TSP‐1 synthesis in other cell types, also reversed the ammonia‐induced TSP‐1 reduction. Likewise, we found a significant decline in TSP‐1 level in cortical astrocytes, as well as a reduction in synaptophysin content in vivo in a rat model of CHE. These findings suggest that TSP‐1 may represent an important therapeutic target for CHE.

     

<Emphasis Type="Italic">atonal</Emphasis>- and <Emphasis Type="Italic">achaete-scute</Emphasis>-related genes in the annelid <Emphasis Type="Italic">Platynereis dumerilii</Emphasis>: insights into the evolution of neural basic-Helix-Loop-Helix genes

Background  

Functional studies in model organisms, such as vertebrates and Drosophila, have shown that basic Helix-loop-Helix (bHLH) proteins have important roles in different steps of neurogenesis, from the acquisition of neural fate to the differentiation into specific neural cell types. However, these studies highlighted many differences in the expression and function of orthologous bHLH proteins during neural development between vertebrates and Drosophila. To understand how the functions of neural bHLH genes have evolved among bilaterians, we have performed a detailed study of bHLH genes during nervous system development in the polychaete annelid, Platynereis dumerilii, an organism which is evolutionary distant from both Drosophila and vertebrates.     

Inhibition of the Transsulfuration Pathway Affects Growth and Feather Follicle Development in Meat-Type Chickens

Cysteine is a nonessential amino acid in poultry nutrition. Poultry diets are deficient in cysteine, but the bird’s cysteine need is met through the transsulfuration pathway (TSP) where homocysteine is converted to cysteine: a process catalyzed by cystathionine β-synthase (CBS) and cystathionine γ-lyase (CTH). Cysteine is also a major component of keratinized protein found in feathers, but the extent to which cysteine is involved in feather and skin development in poultry is unknown. We randomly assigned chicks to control and treatment (control diet plus 100?mg/kg body weight of propargylglycine which is an inhibitor of CTH) diets. The thickness of skin layers, primary feather follicle parameters, growth, and mRNA expression of CBS and CTH were measured. Inhibition of TSP corresponded with the upregulation of liver mRNA of both CBS and CTH and reduction in growth from 35 to 40 days of age. The epidermis thickness, feather follicle length, and diameter were reduced from 10 to 40 days of age. Incorporation of cysteine into keratinized protein may be more sensitive to the level of available cysteine than into nonkeratinized proteins. Thus, disruption of the TSP could affect the thermoregulatory ability of the bird.     

Flrt2 and Flrt3 have overlapping and non-overlapping expression during craniofacial development

Craniofacial morphogenesis is a complex multi-step process that involves numerous biological processes to coordinate the growth, proliferation, migration, and subsequent differentiation of the cranial neural crest cells. Members of the Fibronectin Leucine-Rich Transmembrane (Flrt) gene family have been previously reported to be widely expressed in the developing embryo. We mapped the expression of Flrt2 and Flrt3 at critical stages of craniofacial development and found that, during early craniofacial development, Flrt2 was highly expressed initially in the cranial neural crest cells and Flrt3 in the midbrain. Later both genes were expressed in the developing pharyngeal region. Flrt2 expression predominated in the neural crest-derived mesenchyme in the medial aspect of the developing frontonasal region in close relationships with the expression of Fgfr2, Shh, and Msx1, three genes shown previously to play critical roles in craniofacial development. Flrt2 was also present in the vomero-nasal organ, mandibular primodia, and the posterior aspects of the unfused and fused secondary palatal shelves. Flrt3, however, had a more restrictive expression, being present in the mesenchyme underlying the ectoderm of the medial nasal process and in the mandibular primordium and in regions undergoing outgrowth, in a pattern that overlapped with Bmp4 expression. Both Flrt2 and Flrt3 were later found to be present at sites of epithelial–mesenchymal interactions such as the developing tooth buds, hair follicles, and eye. Together the data suggested important roles for Flrt2 and Flrt3 in mediating events such as NCC migration, chondrogenesis and epithelial–mesenchymal interactions during craniofacial development.     

Entropic analysis and incremental synthesis of multilayered feedforward neural networks

Neural network architecture optimization is often a critical issue, particularly when VLSI implementation is considered. This paper proposes a new minimization method for multilayered feedforward ANNs and an original approach to their synthesis, both based on the analysis of the information quantity (entropy) flowing through the network. A layer is described as an information filter which selects the relevant characteristics until the complete classification is performed. The basic incremental synthesis method, including the supervised training procedure, is derived to design application-tailored neural paradigms with good generalization capability.