Transfer Functions in Artificial Neural Networks - A Simulation-Based Tutorial

Artificial neural networks are based on computational units that resemble basic information processing properties of biological neurons in an abstract and simplified manner. Generally, these formal neurons model an input-output behaviour as it is also often used to characterize biological neurons. The neuron is treated as a black box; spatial extension and temporal dynamics present in biological neurons are most often neglected. Even though artificial neurons are simplified, they can show a variety of input-output relations, depending on the transfer functions they apply. This unit on transfer functions provides an overview of different transfer functions and offers a simulation that visualizes the input-output behaviour of an artificial neuron depending on the specific combination of transfer functions.

Tree species and diversity effects on soil water seepage in a tropical plantation

Plant diversity has been shown to influence the water cycle of forest ecosystems by differences in water consumption and the associated effects on groundwater recharge. However, the effects of biodiversity on soil water fluxes remain poorly understood for native tree species plantations in the tropics. Therefore, we estimated soil water fluxes and assessed the effects of tree species and diversity on these fluxes in an experimental native tree species plantation in Sardinilla (Panama). The study was conducted during the wet season 2008 on plots of monocultures and mixtures of three or six tree species. Rainfall and soil water content were measured and evapotranspiration was estimated with the Penman-Monteith equation. Soil water fluxes were estimated using a simple soil water budget model considering water input, output, and soil water and groundwater storage changes and in addition, were simulated using the physically based one-dimensional water flow model Hydrus-1D. In general, the Hydrus simulation did not reflect the observed pressure heads, in that modeled pressure heads were higher compared to measured ones. On the other hand, the results of the water balance equation (WBE) reproduced observed water use patterns well. In monocultures, the downward fluxes through the 200 cm-depth plane were highest below Hura crepitans (6.13 mm day−1) and lowest below Luehea seemannii (5.18 mm day−1). The average seepage rate in monocultures (±SE) was 5.66 ± 0.18 mm day−1, and therefore, significantly higher than below six-species mixtures (5.49 ± 0.04 mm day−1) according to overyielding analyses. The three-species mixtures had an average seepage rate of 5.63 ± 0.12 mm day−1 and their values did not differ significantly from the average values of the corresponding species in monocultures. Seepage rates were driven by the transpiration of the varying biomass among the plots (r = 0.61, p = 0.017). Thus, a mixture of trees with different growth rates resulted in moderate seepage rates compared to monocultures of either fast growing or slow growing tree species. Our results demonstrate that tree-species specific biomass production and tree diversity are important controls of seepage rates in the Sardinilla plantation during the wet season.

Focal white matter changes in spasmodic dysphonia: a combined diffusion tensor imaging and neuropathological study.

Spasmodic dysphonia is a neurological disorder characterized by involuntary spasms in the laryngeal muscles during speech production. Although the clinical symptoms are well characterized, the pathophysiology of this voice disorder is unknown. We describe here, for the first time to our knowledge, disorder-specific brain abnormalities in these patients as determined by a combined approach of diffusion tensor imaging (DTI) and postmortem histopathology. We used DTI to identify brain changes and to target those brain regions for neuropathological examination. DTI showed right-sided decrease of fractional anisotropy in the genu of the internal capsule and bilateral increase of overall water diffusivity in the white matter along the corticobulbar/corticospinal tract in 20 spasmodic dysphonia patients compared to 20 healthy subjects. In addition, water diffusivity was bilaterally increased in the lentiform nucleus, ventral thalamus and cerebellar white and grey matter in the patients. These brain changes were substantiated with focal histopathological abnormalities presented as a loss of axonal density and myelin content in the right genu of the internal capsule and clusters of mineral depositions, containing calcium, phosphorus and iron, in the parenchyma and vessel walls of the posterior limb of the internal capsule, putamen, globus pallidus and cerebellum in the postmortem brain tissue from one patient compared to three controls. The specificity of these brain abnormalities is confirmed by their localization, limited only to the corticobulbar/corticospinal tract and its main input/output structures. We also found positive correlation between the diffusivity changes and clinical symptoms of spasmodic dysphonia (r = 0.509, P = 0.037). These brain abnormalities may alter the central control of voluntary voice production and, therefore, may underlie the pathophysiology of this disorder.

Internet-based exercises and mini-exams for production and operations management

The purpose of the internet-based teachware mySCM is that students of economics, informatics and industrial engineering get familiar with quantitative methods for supply chain management. Input-output-relationships of various optimization methods can be detected by sampling input values, parameters, and alternative methods for the same problem. Students can gain extra benefits by passing so-called mini-exams that motivate active learning. mySCM can be used for free, round-the-clock, and any place where access to the Internet is available.

Measuring the promise of open data: Development of the Impact Monitoring Framework

It is a challenge to measure the impact of releasing data to the public since the effects may not be directly linked to particular open data activities or substantial impact may only occur several years after publishing the data. This paper proposes a framework to assess the impact of releasing open data by applying the Social Return on Investment (SROI) approach. SROI was developed for organizations intended to generate social and environmental benefits thus fitting the purpose of most open data initiatives. We link the four steps of SROI (input, output, outcome, impact) with the 14 high-value data categories of the G8 Open Data Charter to create a matrix of open data examples, activities, and impacts in each of the data categories. This Impact Monitoring Framework helps data providers to navigate the impact space of open data laying out the conceptual basis for further research.

The European Registry for Patients with Mechanical Circulatory Support (EUROMACS): first annual report

The European Registry for Patients with Mechanical Circulatory Support (EUROMACS) was founded on 10 December 2009 with the initiative of Roland Hetzer (Deutsches Herzzentrum Berlin, Berlin, Germany) and Jan Gummert (Herz- und Diabeteszentrum Nordrhein-Westfalen, Bad Oeynhausen, Germany) with 15 other founding international members. It aims to promote scientific research to improve care of end-stage heart failure patients with ventricular assist device or a total artificial heart as long-term mechanical circulatory support. Likewise, the organization aims to provide and maintain a registry of device implantation data and long-term follow-up of patients with mechanical circulatory support. Hence, EUROMACS affiliated itself with Dendrite Clinical Systems Ltd to offer its members a software tool that allows input and analysis of patient clinical data on a daily basis. EUROMACS facilitates further scientific studies by offering research groups access to any available data wherein patients and centres are anonymized. Furthermore, EUROMACS aims to stimulate cooperation with clinical and research institutions and with peer associations involved to further its aims. EUROMACS is the only European-based Registry for Patients with Mechanical Circulatory Support with rapid increase in institutional and individual membership. Because of the expeditious data input, the European Association for Cardiothoracic Surgeons saw the need to optimize the data availability and the significance of the registry to improve care of patients with mechanical circulatory support and its potential contribution to scientific intents; hence, the beginning of their alliance in 2012. This first annual report is designed to provide an overview of EUROMACS' structure, its activities, a first data collection and an insight to its scientific contributions.

Shadow Profit Maximization and a Generalized Measure of Inefficiency

Determining the profit maximizing input-output bundle of a firm requires data on prices. This paper shows how endogenously determined shadow prices can be used in place of actual prices to obtain the optimal input-output bundle where the firm.s shadow profit is maximized. This approach amounts to an application of the Weak Axiom of Profit Maximization (WAPM) formulated by Varian (1984) based on shadow prices rather than actual prices. At these prices the shadow profit of a firm is zero. Thus, the maximum profit that could have been attained at some other input-output bundle is a measure of the inefficiency of the firm. Because the benchmark input-output bundle is always an observed bundle from the data, it can be determined without having to solve any elaborate programming problem. An empirical application to U.S. airlines data illustrates the proposed methodology.

Role of entorhinal cortical plasticity in spatial and contextual memory

It is well accepted that the hippocampus (HIP) is important for spatial and contextual memories, however, it is not clear if the entorhinal cortex (EC), the main input/output structure for the hippocampus, is also necessary for memory storage. Damage to the EC in humans results in memory deficits. However, animal studies report conflicting results on whether the EC is necessary for spatial and contextual memory. Memory consolidation requires gene expression and protein synthesis, mediated by signaling cascades and transcription factors. Extracellular-signal regulated kinase (ERK) cascade activity is necessary for long-term memory in several tasks, including those that test spatial and contextual memory. In this work, we explore the role of ERK-mediated plasticity in the EC on spatial and contextual memory. ^ To evaluate this role, post-training infusions of reversible pharmacological inhibitors specific for the ERK cascade that do not affect normal neuronal activity were targeted directly to the EC of awake, behaving animals. This technique provides spatial and temporal control over the inhibition of the ERK cascade without affecting performance during training or testing. Using the Morris water maze to study spatial memory, we found that ERK inhibition in the EC resulted in long-term memory deficits consistent with a loss of spatial strategy information. When animals were allowed to learn and consolidate a spatial strategy for solving the task prior to training and ERK inhibition, the deficit was alleviated. To study contextual memory, we trained animals in a cued fear-conditioning task and saw an increase in the activation of ERK in the EC 90 minutes following training. ERK inhibition in the EC over this time point, but not at an earlier time point, resulted in increased freezing to the context, but not to the tone, during a 48-hour retention test. In addition, animals froze maximally at the time the shock was given during training; similar to naïve animals receiving additional training, suggesting that ERK-mediated plasticity in the EC normally suppresses the temporal nature of the freezing response. These findings demonstrate that plasticity in the EC is necessary for both spatial and contextual memory, specifically in the retention of behavioral strategies. ^

Proceedings of the 23rd Annual Biochemical Engineering Symposium

The 23rd Annual Biochemical Engineering Symposium was held at the University of Oklahoma on April 17, 1993. The objectives of the symposium were to provide 1) a forum for informal discussion of biochemical engineering research being carried at the participating universities and 2) an opportunity for students to present and publish their work. Thirteen papers presented at the symposium are included in the proceedings. Because final publication usually takes place in refereed journals, the articles included here are typically brief and often cover work in progress. The program of the symposium and a list of participants are included in the proceedings. ContentsA Low-Cost Bioreactor Strategy for RNA Synthesis, H. Anthony Marble, Eleni Chrisikos, and Robert H. Davis Development of a CELSS Bioreactor: Oxygen Transfer and Micromixing in Parabolic Flight, P.E. Villeneuve, K.S. Wenger, B.G. Thompson, T. Kedar, and E.H. Dunlop Scale-up of Dexter Murine Bone Marrow Cultures Utilizing a Three-Dimensional Fiberglass Support Matrix, John G. Highfill, Paul Todd, Steve Haley, and Dhinaker Kompala Modeling and Estimation of States of Recombinant Fermentations Using Nonlinear Input/Output Models, Vicotr M. Saucedo and M. Nazmul Karim Deadent Microfiltration of Bovine Serum Albumin Suspension Through Yeast Cake Layers and Assymetric Polymeric Membranes, Naveen Arora and Robert H. Davis Monitoring the Fate of Toluene and Phenol in the Rhizosphere, N. Muralidharan, Lawrence C. Davis, and Larry E. Erickson Hydrodynamic Motions Associated with Bubble Coalescence and Breakup, T.Y. Yiin, L.A. Glasgow, and L.E. Erickson Expression and Purification of a-Human Atrial Natriuretic Peptide in Escherichia coli by Fusion with L-Asparaginase, Nien-Tung Ma and Roger G. Harrison High Pressure Crystallization of Proteins, Mungara V. Saikumar, Charles E. Glatz, and Maurice A. Larson Structure/Function Relationships in the Catalytic and Starch Binding Domains of Glucoamylase, Pedro M. Coutinho, Clark Ford, Peter J. Reilly Cellular Responses of Insect Cell Spodoptera frugiperda to Environmental Stresses, Paul Yeh, Grace Y. Sun, Gary A. Weisman, Rakesh Bajpai A Novel Approach to Understanding the Antimicrobial Activity of Peptides, Naveen Pathak, Marie-Helene Janna, Gael Ruche, David McCarthy, and Roger Harrison Mass Transfer in the Bioremediation of Soils Contaminated with Trapped Non-Aqueous Phase Liquids, Xiaoqing Yang, Larry E. Jacobson, and L.T. Fan

Business Cycles of Non-mono-cultural Developing Economies: The Case of ASEAN Countries

Based on analyses of actual data, we reveal that many Asian developing economies own economic structural features of "non-mono-cultural economy" and the "large primary good sector", which have not been discussed in developing economies RBC literature. We also examine the input-output tables to develop a model reflecting actual developing economies' structures. Referring to the analyses, we construct RBC models of ASEAN countries. Based on the model, we find that approximately half of GDP volatility is attributable to domestic productivity shocks, and the remaining half is attributable to price shocks.

Application of factor decomposition techniques to vertical specialisation measurements

The increasing importance of vertical specialisation (VS) trade has been a notable feature of rapid economic globalisation and regional integration. In an attempt to understand countries’ depth of participation in global production chains, many Input-Output based VS indicators have been developed. However, most of them focus on showing the overall magnitude of a country’s VS trade, rather than explaining the roles that specific sectors or products play in VS trade and what factors make the VS change over time. Changes in vertical specialisation indicators are, in fact, determined by mixed and complex factors such as import substitution ratios, types of exported goods and domestic production networks. In this paper, decomposition techniques are applied to VS measurement based on the OECD Input-Output database. The decomposition results not only help us understand the structure of VS at detailed sector and product levels, but also show us the contributions of trade dependency, industrial structures of foreign trade and domestic production system to a country’s vertical specialisation trade.

Chapter 2 On the usage of the measurements of geographical concentration and specialization with areal data

Measures have been developed to understand tendencies in the distribution of economic activity. The merits of these measures are in the convenience of data collection and processing. In this interim report, investigating the property of such measures to determine the geographical spread of economic activities, we summarize the merits and limitations of measures, and make clear that we must apply caution in their usage. As a first trial to access areal data, this project focus on administrative areas, not on point data and input-output data. Firm level data is not within the scope of this article. The rest of this article is organized as follows. In Section 2, we touch on the the limitations and problems associated with the measures and areal data. Specific measures are introduced in Section 3, and applied in Section 4. The conclusion summarizes the findings and discusses future work.

Assessing agglomeration economies in the Yangzi River Delta, China : a bayesian spatial econometric approach

This paper estimates the elasticity of labor productivity with respect to employment density, a widely used measure of the agglomeration effect, in the Yangtze River Delta, China. A spatial Durbin model is presented that makes explicit the influences of spatial dependence and endogeneity bias in a very simple way. Results of Bayesian estimation using the data of the year 2009 indicate that the productivity is influenced by factors correlated with density rather than density itself and that spatial spillovers of these factors of agglomeration play a significant role. They are consistent with the findings of Ke (2010) and Artis, et al. (2011) that suggest the importance of taking into account spatial dependence and hitherto omitted variables.

Agglomeration and firm-level productivity : a Bayesian spatial approach

This paper estimates the impact of industrial agglomeration on firm-level productivity in Chinese manufacturing sectors. To account for spatial autocorrelation across regions, we formulate a hierarchical spatial model at the firm level and develop a Bayesian estimation algorithm. A Bayesian instrumental-variables approach is used to address endogeneity bias of agglomeration. Robust to these potential biases, we find that agglomeration of the same industry (i.e. localization) has a productivity-boosting effect, but agglomeration of urban population (i.e. urbanization) has no such effects. Additionally, the localization effects increase with educational levels of employees and the share of intermediate inputs in gross output. These results may suggest that agglomeration externalities occur through knowledge spillovers and input sharing among firms producing similar manufactures.