Simslerche8546

This paper presents the multistability analysis and associative memory of neural networks (NNs) with Morita-like activation functions. In order to seek larger memory capacity, this paper proposes Morita-like activation functions. In a weakened condition, this paper shows that the NNs with n-neurons have (2m+1)n equilibrium points (Eps) and (m+1)n of them are locally exponentially stable, where the parameter m depends on the Morita-like activation functions, called Morita parameter. Also the attraction basins are estimated based on the state space partition. Moreover, this paper applies these NNs into associative memories (AMs). Compared with the previous related works, the number of Eps and AM's memory capacity are extensively increased. The simulation results are illustrated and some reliable associative memories examples are shown at the end of this paper.Neural networks have become standard tools in the analysis of data, but they lack comprehensive mathematical theories. For example, there are very few statistical guarantees for learning neural networks from data, especially for classes of estimators that are used in practice or at least similar to such. In this paper, we develop a general statistical guarantee for estimators that consist of a least-squares term and a regularizer. We then exemplify this guarantee with ℓ1-regularization, showing that the corresponding prediction error increases at most logarithmically in the total number of parameters and can even decrease in the number of layers. Our results establish a mathematical basis for regularized estimation of neural networks, and they deepen our mathematical understanding of neural networks and deep learning more generally.Robustness of deep neural networks is a critical issue in practical applications. In the general case of feed-forward neural networks (including convolutional deep neural network architectures), under random noise attacks, we propose to study the probability that the output of the network deviates from its nominal value by a given threshold. We derive a simple concentration inequality for the propagation of the input uncertainty through the network using the Cramer-Chernoff method and estimates of the local variation of the neural network mapping computed at the training points. We further discuss and exploit the resulting condition on the network to regularize the loss function during training. Finally, we assess the proposed tail probability estimates empirically on various public datasets and show that the observed robustness is very well estimated by the proposed method.The brain is able to calculate the distance and direction to the desired position based on grid cells. Extensive neurophysiological studies of rodent navigation have postulated the grid cells function as a metric for space, and have inspired many computational studies to develop innovative navigation approaches. Furthermore, grid cells may provide a general encoding scheme for high-order nonspatial information. Built upon existing neuroscience and machine learning work, this paper provides theoretical clarity on that the grid cell population codes can be taken as a metric for space. The metric is generated by a shift-invariant positive definite kernel via kernel distance method and embeds isometrically in a Euclidean space, and the inner product of the grid cell population code exponentially converges to the kernel. We also provide a method to learn the distribution of grid cell population efficiently. Grid cells, as a scalable position encoding method, can encode the spatial relationships of places and enable grid cells to outperform place cells in navigation. Further, we extend the grid cell to images encoding and find that grid cells embed images into a mental map, where geometric relationships are conceptual relationships of images. The theoretical model and analysis would contribute to establishing the grid cell code as a generic coding scheme for both spatial and conceptual spaces, and is promising for a multitude of problems across spatial cognition, machine learning and semantic cognition.While chronic visual symptom complaints are common among Veterans with a history of mild traumatic brain injury (mTBI), research is still ongoing to characterize the pattern of visual deficits that is most strongly associated with mTBI and specifically, the impact of blast-related mTBI on visual functioning. One area that has not been well explored is the potential impact of blast mTBI on refractive error. While myopic shifts have been documented following head injuries in civilian populations, posttraumatic myopic shifts have not been explored in participants with military mTBI. This study investigated the impact of blast mTBIs on a range of visual function measures including distance acuity and refractive error, in a well-characterized cohort of thirty-one Post-9/11 veterans for whom detailed clinical interviews regarding military and TBI history were available. Seventeen participants had a history of blast-related mTBI (blast mTBI + group) while 14 did not (blast mTBI- group). Results show an increased frequency of convergence insufficiency and myopia in the blast mTBI + group relative to the blast mTBI- group. Linear regression analyses further show that deficits in distance acuity and refractive error are associated with the number of blast mTBIs during military service but not the number of non-blast mTBIs or the number of lifetime non-blast TBIs and cannot be accounted for by PTSD. These results are consistent with long-lasting damage following blast mTBI to subcortical visual structures that support both vergence movements and the accommodative functions needed to see clearly objects at varying distances.The physiological disorders in humans resulting from the excess dietary intake of manganese (Mn) via whole-grain food has attracted considerable attention. However, the speciation and bioavailability of Mn in wheat grains and their response to different phosphorus (P) fertilization rates are still unclear. In the current study, using a long-term field trial with P application rates of 0, 21.8, 43.6, 65.5 and 87.3 kg/ha, we examined changes in the concentration, distribution, and speciation of Mn of wheat grains using synchrotron-based X-ray fluorescence microscopy and X-ray absorption spectroscopy. The total Mn concentration in grains was found to be increased by phosphorus fertilization, especially in embryo in the form of Mn(II), but this phosphorus fertilization also decreased Mn concentrations in the nucellar projection. In this study, the speciation of Mn in different wheat grain tissues was examined, and results indicate that in calcareous soils, high rates of P fertilizers can increase Mn concentrations in wheat grain, including Mn which is likely to be of high bioavailability, and thus may increase the risk for human to expose to high Mn intake via whole-grain food.Arsenic (As), mercury (Hg), and copper (Cu) are among the major historical and contemporary metal pollutants linked to global anthropogenic activities. Enterococcus have been considered indicators of fecal pollution and antibiotic resistance for years, but its largely underexplored metallome precludes understanding their role as metal pollution bioindicators as well. Our goal was to determine the occurrence, diversity, and phenotypes associated with known acquired genes/operons conferring tolerance to As, Hg or Cu among Enterococcus and to identify their genetic context (381 field isolates from diverse epidemiological and genetic backgrounds; 3547 enterococcal genomes available in databases representing a time span during 1900-2019). Genes conferring tolerance to As (arsA), Hg (merA) or Cu (tcrB) were used as biomarkers of widespread metal tolerance operons. Different variants of metal tolerance (MeT) genes (13 arsA, 6 merA, 1 tcrB) were more commonly recovered from the food-chain (arsA, tcrB) or humans (merAvironments by anthropogenic activities.The alteration of rare earth elements (REEs) biogeochemical cycles has increased the potential effects related to their environmental exposure in a one-health perspective. Cerium (Ce), gadolinium (Gd), lanthanum (La), and neodymium (Nd) are frequently related to technological applications and their environmental concentrations are already in the μg/kg - mg/kg (i.e., or L) range depending on the considered matrices. The effect of Ce, Gd, La, and Nd was investigated in a simulated AMD (0.01-10.22 mg/L) at pH 4 and 6 considering a battery of photosynthetic organisms (Raphidocelis subcapitata, Lepidium sativum, and Vicia faba) according to a multiple-endpoint approach (growth inhibition, germination index, and mutagenicity). According to modelled chemical speciation, the considered elements were mostly in the trivalent free form (86-88%) at pH 4. Gd, La, and Nd exerted the most relevant toxic effect at pH 4. The pH 6 scenario evidenced a reduction in REEs toxicity level. Mutagenicity was detected only at pH 4 by Gd (up to 3-fold compared to negative controls), La and Nd, while Ce did not show any adverse effect. Toxic effects due to Ce, Gd, La, and Nd can be reduced by controlling the pH, but several gaps of knowledge still remain about their uptake and trophic transfer, and long-term effects on targeted species.Fumonisin B1 (FB1) is a contaminant that commonly present in the global environment, especially in food and feed. Epidemiologic studies have shown that esophageal cancer is associated with fumonisin toxicity. However, the molecular mechanism of FB1-induced esophageal cancer is unclear. In this research, the molecular mechanism of FB1-induced cell carcinogenesis in human esophageal epithelial cells line (HEEC) was explored. We found that FB1 (0.3125-5 μM) could promote cell proliferation, and the same phenomenon was found in a 3D cell model. FB1 could also accelerate cell migration. The expression levels of DNA damage markers were significantly increased after FB1 exposure. Meanwhile, the expression levels of cell cycle-regulated proteins and cancer-related genes were abnormal. Furthermore, FB1 significantly upregulated the histone deacetylase (HDAC) expression and activated the phosphoinositide 3 kinase (PI3K)/protein kinase B (Akt) signalling pathway. The HDAC inhibitor trichostatin A (TSA) could repressed FB1-promoted cell proliferation and abnormal phenomenon induced by FB1. Moreover, myriocin (ISP-1) could relieve FB1-enhanced HDAC expression and cell proliferation, which implied that ISP-1 may be used to block the fumonisin toxicity in the future. Our findings suggested that the HDAC/PI3K/Akt signalling pathway is a novel mechanism for FB1-induced cell carcinogenesis in HEEC and provided new ideas for the prevention and control of fumonisin toxicity, subsequently avoiding adverse effects on the ecosystem and human health.Prevalence of nitrate in different aquifer systems is a growing environmental and public health concern. Efforts were made for the first-time to achieve a higher accuracy in health risks characterization associated with the nitrate in groundwater of the diverse aquifer systems on the residents of a semi-arid rural tract of Lower Ganga Basin using Monte Carlo Simulations and Sobol Sensitivity analyses. The nitrate levels in groundwater varied between 0 and 508.3 mg/L with a mean of 19.79 ± 32.78 mg/L and 0-435.0 mg/L with a mean of 24.44 ± 35.15 mg/L during the pre-monsoon and post-monsoon periods, respectively. About 847.12-1000.25 km2 area of the survey area (total area 4545 km2) exhibited nitrate concentrations (C) > the pre-intervention limits (45-50 mg/L). Minor populations, especially the infants from the granite gneiss, Rajmahal traps, laterite, recent alluvial and old alluvial aquifer zones under the Central Tendency Exposure (CTE) condition and all the aquifer zones (including the Gondwana supergroup aquifer zone) under Reasonable Maximum Exposure (RME) scenarios, were characterized as being at high risks of methemoglobinemia, primarily due to ingestion of untreated nitrate contaminated groundwater.