Compact yet efficient hardware implementation of artificial neural networks with customized topology

There are several neural network implementations using either software, hardware-based or a hardware/software co-design. This work proposes a hardware architecture to implement an artificial neural network (ANN), whose topology is the multilayer perceptron (MLP). In this paper, we explore the parallelism of neural networks and allow on-the-fly changes of the number of inputs, number of layers and number of neurons per layer of the net. This reconfigurability characteristic permits that any application of ANNs may be implemented using the proposed hardware. In order to reduce the processing time that is spent in arithmetic computation, a real number is represented using a fraction of integers. In this way, the arithmetic is limited to integer operations, performed by fast combinational circuits. A simple state machine is required to control sums and products of fractions. Sigmoid is used as the activation function in the proposed implementation. It is approximated by polynomials, whose underlying computation requires only sums and products. A theorem is introduced and proven so as to cover the arithmetic strategy of the computation of the activation function. Thus, the arithmetic circuitry used to implement the neuron weighted sum is reused for computing the sigmoid. This resource sharing decreased drastically the total area of the system. After modeling and simulation for functionality validation, the proposed architecture synthesized using reconfigurable hardware. The results are promising.     

A stochastic local search algorithm for constrained continuous global optimization

This paper presents a new stochastic local search algorithm known as feasible–infeasible search procedure (FISP) for constrained continuous global optimization. The proposed procedure uses metaheuristic strategies for combinatorial optimization as well as combined strategies for exploring continuous spaces, which are applied to an efficient process in increasingly refined neighborhoods of current points. We show effectiveness and efficiency of the proposed procedure on a standard set of 13 well‐known test problems. Furthermore, we compare the performance of FISP with SNOPT (sparse nonlinear optimizer) and with few successful existing stochastic algorithms on the same set of test problems.     

Brazilian articles in international journals on Limnology

Summary  We assessed the contribution of Brazilian limnologists (freshwater ecologists) in international journals in the period 1970-2004. Brazilian contribution was low and regular in the 1970's, but increased steeply after 1980 with no signs of stabilization until the present. Articles authored by Brazilians tend to be less cited than articles authored by non-Brazilians, although this difference is reduced in co-authored articles with international researchers.  Brazilian articles are not distributed homogenously among the sub-areas of Limnology, but present some biases that can be explained by intellectual legacy. Brazil has invested since the 1970's in establishing postgraduate courses in Brazil and in the last years has turned the focus to a better qualification of these courses. We believe these are the main reasons for the conspicuous development of Brazilian Limnology.     

Detection and classification of highway lanes using vehicle motion trajectories

Intelligent vision-based traffic surveillance systems are assuming an increasingly important role in highway monitoring and road management schemes. This paper describes a low-level object tracking system that produces accurate vehicle motion trajectories that can be further analyzed to detect lane centers and classify lane types. Accompanying techniques for indexing and retrieval of anomalous trajectories are also derived. The predictive trajectory merge-and-split algorithm is used to detect partial or complete occlusions during object motion and incorporates a Kalman filter that is used to perform vehicle tracking. The resulting motion trajectories are modeled using variable low-degree polynomials. A K-means clustering technique on the coefficient space can be used to obtain approximate lane centers. Estimation bias due to vehicle lane changes can be removed using robust estimation techniques based on Random Sample Consensus (RANSAC). Through the use of nonmetric distance functions and a simple directional indicator, highway lanes can be classified into one of the following categories: entry, exit, primary, or secondary. Experimental results are presented to show the real-time application of this approach to multiple views obtained by an uncalibrated pan-tilt-zoom traffic camera monitoring the junction of two busy intersecting highways.     

Quantum Phase Transition in the BCS-to-BEC Evolution of p-wave Fermi Gases

We discuss the possibility of a quantum phase transition in ultra-cold spin-polarized Fermi gases which exhibit a p-wave Feshbach resonance. We show that when fermionic atoms form a condensate that can be externally tuned between the BCS and BEC limits, the zero temperature compressibility and the spin susceptibility of the fermionic gas are non-analytic functions of the two-body bound state energy. This non-analyticity is due to a massive rearrangement of the momentum distribution in the ground state of the system. Furthermore, we show that the low temperature superfluid density is also non-analytic and exhibits a dramatic change in behavior when the critical value of the bound state energy is crossed.     

NAROAS: a neural network-based advanced operator support system for the assessment of systems reliability

We have developed and implemented a computerized reliability monitoring system for nuclear power plant applications, based on a neural network. The developed computer program is a new tool related to operator decision support systems, in case of component failures, for the determination of test and maintenance policies during normal operation or to follow an incident sequence in a nuclear power plant. The NAROAS (Neural Network Advanced Reliability Advisory System) computer system has been developed as a modularized integrated system in a C++ Builder environment, using a Hopfield neural network instead of fault trees, to follow and control the different system configurations, for interventions as quickly as possible at the plant. The observed results are comparable and similar to those of other computer system results. As shown, the application of this neural network contributes to the state of the art of risk monitoring systems by turning it easier to perform online reliability calculations in the context of probabilistic safety assessments of nuclear power plants.     

Dynamic multi-commodity capacitated facility location: a mathematical modeling framework for strategic supply chain planning

In this paper, we focus on the strategic design of supply chain networks. We propose a mathematical modeling framework that captures many practical aspects of network design problems simultaneously but which have not received adequate attention in the literature. The aspects considered include: dynamic planning horizon, generic supply chain network structure, external supply of materials, inventory opportunities for goods, distribution of commodities, facility configuration, availability of capital for investments, and storage limitations. Moreover, network configuration decisions concerning the gradual relocation of facilities over the planning horizon are considered. To cope with fluctuating demands, capacity expansion and reduction scenarios are also analyzed as well as modular capacity shifts. The relation of the proposed modeling framework with existing models is discussed. For problems of reasonable size, we report on our computational experience with standard mathematical programming software. In particular, useful insights on the impact of various factors on network design decisions are provided.     

t-Spanners for metric space searching

The problem of Proximity Searching in Metric Spaces consists in finding the elements of a set which are close to a given query under some similarity criterion. In this paper we present a new methodology to solve this problem, which uses a t-spanner G′(V, E) as the representation of the metric database. A t-spanner is a subgraph G′(V, E) of a graph G(V, A), such that E  A and G′ approximates the shortest path costs over G within a precision factor t.

Our key idea is to regard the t-spanner as an approximation to the complete graph of distances among the objects, and to use it as a compact device to simulate the large matrix of distances required by successful search algorithms such as AESA. The t-spanner properties imply that we can use shortest paths over G′ to estimate any distance with bounded-error factor t.

For this sake, several t-spanner construction, updating, and search algorithms are proposed and experimentally evaluated. We show that our technique is competitive against current approaches. For example, in a metric space of documents our search time is only 9% over AESA, yet we need just 4% of its space requirement. Similar results are obtained in other metric spaces.

Finally, we conjecture that the essential metric space property to obtain good t-spanner performance is the existence of clusters of elements, and enough empirical evidence is given to support this claim. This property holds in most real-world metric spaces, so we expect that t-spanners will display good behavior in most practical applications. Furthermore, we show that t-spanners have a great potential for improvements.     

Towards the Search for Potential Biomarkers in Osteosarcoma: State-of-the-Art and Translational Expectations

Osteosarcoma represents a rare cause of cancer in the general population, accounting for <1% of malignant neoplasms globally. Nonetheless, it represents the main cause of malignant bone neoplasm in children, adolescents and young adults under 20 years of age. It also presents another peak of incidence in people over 50 years of age and is associated with rheumatic diseases. Numerous environmental risk factors, such as bone diseases, genetics and a history of previous neoplasms, have been widely described in the literature, which allows monitoring a certain group of patients. Diagnosis requires numerous imaging tests that make it possible to stratify both the local involvement of the disease and its distant spread, which ominously determines the prognosis. Thanks to various clinical trials, the usefulness of different chemotherapy regimens, radiotherapy and surgical techniques with radical intent has now been demonstrated; these represent improvements in both prognosis and therapeutic approaches. Osteosarcoma patients should be evaluated in reference centres by multidisciplinary committees with extensive experience in proper management. Although numerous genetic and rheumatological diseases and risk factors have been described, the use of serological, genetic or other biomarkers has been limited in clinical practice compared to other neoplasms. This limits both the initial follow-up of these patients and screening in populations at risk. In addition, we cannot forget that the diagnosis is mainly based on the direct biopsy of the lesion and imaging tests, which illustrates the need to study new diagnostic alternatives. Therefore, the purpose of this study is to review the natural history of the disease and describe the main biomarkers, explaining their clinical uses, prognosis and limitations.     

CD73-Mediated Formation of Extracellular Adenosine Is Responsible for Adenosine A2A Receptor-Mediated Control of Fear Memory and Amygdala Plasticity

Adenosine A_2A receptors (A_2AR) control fear memory and the underlying processes of synaptic plasticity in the amygdala. In other brain regions, A_2AR activation is ensured by ATP-derived extracellular adenosine formed by ecto-5′-nucleotidase or CD73. We now tested whether CD73 is also responsible to provide for the activation of A_2AR in controlling fear memory and amygdala long-term potentiation (LTP). The bilateral intracerebroventricular injection of the CD73 inhibitor αβ-methylene ADP (AOPCP, 1 nmol/ventricle/day) phenocopied the effect of the A_2AR blockade by decreasing the expression of fear memory, an effect disappearing in CD73-knockout (KO) mice and in forebrain neuronal A_2AR-KO mice. In the presence of PPADS (20 μM) to eliminate any modification of ATP/ADP-mediated P2 receptor effects, both AOPCP (100 μM) and the A_2AR antagonist, {"type":"entrez-protein","attrs":{"text":"SCH58261","term_id":"1052882304","term_text":"SCH58261"}}SCH58261 (50 nM), decreased LTP magnitude in synapses of projection from the external capsula into the lateral amygdala, an effect eliminated in slices from both forebrain neuronal A_2AR-KO mice and CD73-KO mice. These data indicate a key role of CD73 in the process of A_2AR-mediated control of fear memory and underlying synaptic plasticity processes in the amygdala, paving the way to envisage CD73 as a new therapeutic target to interfere with abnormal fear-like emotional processing.