A new Jacobian matrix for optimal learning of single-layer neural networks

This paper investigates the learning of a wide class of single-hidden-layer feedforward neural networks (SLFNs) with two sets of adjustable parameters, i.e., the nonlinear parameters in the hidden nodes and the linear output weights. The main objective is to both speed up the convergence of second-order learning algorithms such as Levenberg-Marquardt (LM), as well as to improve the network performance. This is achieved here by reducing the dimension of the solution space and by introducing a new Jacobian matrix. Unlike conventional supervised learning methods which optimize these two sets of parameters simultaneously, the linear output weights are first converted into dependent parameters, thereby removing the need for their explicit computation. Consequently, the neural network (NN) learning is performed over a solution space of reduced dimension. A new Jacobian matrix is then proposed for use with the popular second-order learning methods in order to achieve a more accurate approximation of the cost function. The efficacy of the proposed method is shown through an analysis of the computational complexity and by presenting simulation results from four different examples.

2-D R-matrix propagation: A large scale electron scattering simulation dominated by the multiplication of dynamically changing matrices

We examine the computational aspects of propagating a global R-matrix, R, across sub-regions in a 2-D plane. This problem originates in the large scale simulation of electron collisions with atoms and ions at intermediate energies. The propagation is dominated by matrix multiplications which are complicated because of the dynamic nature of R, which changes the designations of its rows and columns and grows in size as the propagation proceeds. The use of PBLAS to solve this problem on distributed memory HPC machines is the main focus of the paper.

Ultrafast Laser-Driven Excitation Dynamics in Ne: An Ab Initio Time-Dependent R-Matrix Approach

An attosecond pump-probe scheme that combines the use of a free-electron laser pulse with an ultrashort pulse is applied in order to explore the ultrafast excitation dynamics in Ne. We describe the multielectron dynamics using a new nonperturbative time-dependent R-matrix theory. This theory enables the interaction of ultrashort light fields with multielectron atoms and atomic ions to be determined from first principles. By probing the emission of an inner 2s electron from Ne we are also able to study the bound state population dynamics during the free-electron laser pulse.

Time-dependent R -matrix theory for ultrafast atomic processes

We describe an ab initio nonperturbative time-dependent R-matrix theory for ultrafast atomic processes. This theory enables investigations of the interaction of few-femtosecond and -attosecond pulse lasers with complex multielectron atoms and atomic ions. A derivation and analysis of the basic equations are given, which propagate the atomic wave function in the presence of the laser field forward in time in the internal and external R-matrix regions. To verify the accuracy of the approach, we investigate two-photon ionization of Ne irradiated by an intense laser pulse and compare current results with those obtained using the R-matrix Floquet method and an alternative time-dependent method. We also verify the capability of the current approach by applying it to the study of two-dimensional momentum distributions of electrons ejected from Ne due to irradiation by a sequence of 2 as light pulses in the presence of a 780 nm laser field.

2DRMP: A suite of two-dimensional R-matrix propagation codes

The R-matrix method has proved to be a remarkably stable, robust and efficient technique for solving the close-coupling equations that arise in electron and photon collisions with atoms, ions and molecules. During the last thirty-four years a series of related R-matrix program packages have been published periodically in CPC. These packages are primarily concerned with low-energy scattering where the incident energy is insufficient to ionize the target. In this paper we describe previous term2DRMP,next term a suite of two-dimensional R-matrix propagation programs aimed at creating virtual experiments on high performance and grid architectures to enable the study of electron scattering from H-like atoms and ions at intermediate energies.

Gelation of ionic liquids using a cross-linked poly(ethylene glycol) gel matrix

Conductive ionic liquid -poly(ethylene glycol) (IL-PEG) gels have been prepared by gelation of the hydrophobic ionic liquid 1-hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide [(C(6)mim] [NTf2]) by the cross-linking reaction of disuccinimidylpropyl PEG monomers with four-arm tetraamine PEG cross-linkers. This is the first time that a crosslinked PEG matrix, such as this, has been used to gel nonaqueous solvents. Initial studies screening other ionic liquids as solvents indicate that the gelation of the ionic liquid is both cation and anion dependent with smaller, coordinating cations disrupting or preventing gel formation.

Alpha-1-antitrypsin aerosolized augmentation abrogates neutrophil elastase-induced expression of cathepsin B and matrix metalloprotease 2 in vivo and in vitro

Background: Neutrophil elastase (NE) activity is increased in lung diseases such as a1-antitrypsin (A1AT) deficiency and pneumonia. It has recently been shown to induce expression of cathepsin B and matrix metalloprotease 2 (MMP-2) in vitro and in a mouse model. It is postulated that increased cathepsin B and MMP-2 in acute and chronic lung diseases result from high levels of extracellular NE and that expression of these proteases could be inhibited by A1AT augmentation therapy.

Methods: Cathepsin and MMP activities were assessed in bronchoalveolar lavage (BAL) fluid from patients with A1AT deficiency, pneumonia and control subjects. Macrophages were exposed to BAL fluid rich in free NE from patients with pneumonia following pretreatment with A1AT. MMP-2, cathepsin B, secretory leucoprotease inhibitor (SLPI) and lactoferrin levels were determined in BAL fluid from A1AT-deficient patients before and after aerosolisation of A1AT.

Results: BAL fluid from both patients with pneumonia and those with A1AT deficiency containing free NE had increased cathepsin B and MMP-2 activities compared with BAL fluid from healthy volunteers. The addition of A1AT to BAL fluid from patients with pneumonia greatly reduced NE-induced cathepsin B and MMP-2 expression in macrophages in vitro. A1AT augmentation therapy to A1AT-deficient individuals also reduced cathepsin B and MMP-2 activity in BAL fluid in vivo. Furthermore, A1AT-deficient patients had higher levels of SLPI and lactoferrin after A1AT augmentation therapy.

Conclusion: These findings suggest a novel role for A1AT inhibition of NE-induced upregulation of MMP and cathepsin expression both in vitro and in vivo.

Advanced glycation end products cause increased CCN family and extracellular matrix gene expression in the diabetic rodent retina

Aims/hypothesis: Referred to as CCN, the family of growth factors consisting of cystein-rich protein 61 (CYR61, also known as CCN1), connective tissue growth factor (CTGF, also known as CCN2), nephroblastoma overexpressed gene (NOV, also known as CCN3) and WNT1-inducible signalling pathway proteins 1, 2 and 3 (WISP1, -2 and -3; also known as CCN4, -5 and -6) affects cellular growth, differentiation, adhesion and locomotion in wound repair, fibrotic disorders, inflammation and angiogenesis. AGEs formed in the diabetic milieu affect the same processes, leading to diabetic complications including diabetic retinopathy. We hypothesised that pathological effects of AGEs in the diabetic retina are a consequence of AGE-induced alterations in CCN family expression.

Materials and methods: CCN gene expression levels were studied at the mRNA and protein level in retinas of control and diabetic rats using real-time quantitative PCR, western blotting and immunohistochemistry at 6 and 12 weeks of streptozotocin-induced diabetes in the presence or absence of aminoguanidine, an AGE inhibitor. In addition, C57BL/6 mice were repeatedly injected with exogenously formed AGE to establish whether AGE modulate retinal CCN growth factors in vivo.

Results: After 6 weeks of diabetes, Cyr61 expression levels were increased more than threefold. At 12 weeks of diabetes, Ctgf expression levels were increased twofold. Treatment with aminoguanidine inhibited Cyr61 and Ctgf expression in diabetic rats, with reductions of 31 and 36%, respectively, compared with untreated animals. Western blotting showed a twofold increase in CTGF production, which was prevented by aminoguanidine treatment. In mice infused with exogenous AGE, Cyr61 expression increased fourfold and Ctgf expression increased twofold in the retina.

Conclusions/interpolation: CTGF and CYR61 are downstream effectors of AGE in the diabetic retina, implicating them as possible targets for future intervention strategies against the development of diabetic retinopathy.

Structural change in multipartite entanglement sharing: A random matrix approach

We study the typical entanglement properties of a system comprising two independent qubit environments interacting via a shuttling ancilla. The initial preparation of the environments is modeled using random matrix techniques. The entanglement measure used in our study is then averaged over many histories of randomly prepared environmental states. Under a Heisenberg interaction model, the average entanglement between the ancilla and one of the environments remains constant, regardless of the preparation of the latter and the details of the interaction. We also show that, upon suitable kinematic and dynamical changes in the ancillaenvironment subsystems, the entanglement-sharing structure undergoes abrupt modifications associated with a change in the multipartite entanglement class of the overall system's state. These results are invariant with respect to the randomized initial state of the environments.

Salbutamol up-regulates matrix metalloproteinase-9 in the alveolar space in the acute respiratory distress syndrome.

Objectives: Acute respiratory distress syndrome (ARDS) is characterized by alveolar-capillary barrier damage. Matrix metalloproteinases (MMPs) are implicated in the pathogenesis of ARDS. In the Beta Agonists in Acute Lung Injury Trial, intravenous salbutamol reduced extravascular lung water (EVLW) in patients with ARDS at day 4 but not inflammatory cytokines or neutrophil recruitment. We hypothesized that salbutamol reduces MMP activity in ARDS.

Methods: MMP-1/-2/-3/-7/-8/-9/-12/-13 was measured in supernatants of distal lung epithelial cells, type II alveolar cells, and bronchoalveolar lavage (BAL) fluid from patients in the Beta Agonists in Acute Lung Injury study by multiplex bead array and tissue inhibitors of metalloproteinases (TIMPs)-1/-2 by enzyme-linked immunosorbent assay. MMP-9 protein and activity levels were further measured by gelatin zymography and fluorokine assay.

Measurements and Main Results: BAL fluid MMP-1/-2/-3 declined by day 4, whereas total MMP-9 tended to increase. Unexpectedly, salbutamol augmented MMP-9 activity. Salbutamol induced 33.7- and 13.2-fold upregulation in total and lipocalin-associated MMP-9, respectively at day 4, compared with 2.0- and 1.3-fold increase in the placebo group, p < 0.03. Salbutamol did not affect BAL fluid TIMP-1/-2. Net active MMP-9 was higher in the salbutamol group (4222 pg/mL, interquartile range: 513-7551) at day 4 compared with placebo (151 pg/mL, 124-2108), p = 0.012. Subjects with an increase in BAL fluid MMP-9 during the 4-day period had lower EVLW measurements than those in whom MMP-9 fell (10 vs. 17 mL/kg, p = 0.004): change in lung water correlated inversely with change in MMP-9, r = -.54, p = 0.0296. Salbutamol up-regulated MMP-9 and down-regulated TIMP-1/-2 secretion in vitro by distal lung epithelial cells. Inhibition of MMP-9 activity in cultures of type II alveolar epithelial cells reduced wound healing.

Conclusions: Salbutamol specifically up-regulates MMP-9 in vitro and in vivo in patients with ARDS. Up-regulated MMP-9 is associated with a reduction in EVLW. MMP-9 activity is required for alveolar epithelial wound healing in vitro. Data suggest MMP-9 may have a previously unrecognized beneficial role in reducing pulmonary edema in ARDS by improving alveolar epithelial healing.

The matrix metalloproteinase-3 (MMP-3) 5A/6A promoter polymorphism is not associated with ischaemic heart disease: Analysis employing a family based approach

Matrix metalloproteinase-3 (MMP-3) has been proposed as an important mediator of the atherosclerotic process. The possible role of the functional -1612(.)5A/6A polymorphism of the MMP-3 gene in the susceptibility to ischaemic heart disease (IHD) was investigated in a well-defined Irish population using two recently described family based tests of association. One thousand and twelve individuals from 386 families with at least one member prematurely affected with IHD were genotyped. Using the combined transmission disequilibrium test (TDT)/sib-TDT and the pedigree disequilibrium test (PDT), no association between the MMP-3 -1612 5A/6A polymorphism and IHD was found. Our data demonstrate that, in an Irish population, the MMP-3 -1612 5A/6A polymorphism is not associated with IHD.

Eigenvalues of the time-delay matrix in overlapping resonances

We discuss the properties of the lifetime or the time-delay matrix Q(E) for multichannel scattering, which is related to the scattering matrix S(E) by Q = i?S(dS†/dE). For two overlapping resonances occurring at energies E with widths G(? = 1, 2), with an energy-independent background, only two eigenvalues of Q(E) are proved to be different from zero and to show typical avoided-crossing behaviour. These eigenvalues are expressible in terms of the four resonance parameters (E , G) and a parameter representing the strength of the interaction of the resonances. An example of the strong and weak interaction in an overlapping double resonance is presented for the positronium negative ion. When more than two resonances overlap (? = 1, ..., N), no simple representation of each eigenvalue has been found. However, the formula for the trace of the Q-matrix leads to the expression d(E) = -?arctan[(G/2)/(E - E)] + d(E) for the eigenphase sum d(E) and the background eigenphase sum d(E), in agreement with the known form of the state density. The formulae presented in this paper are useful in a parameter fitting of overlapping resonances. © 2006 IOP Publishing Ltd.

Fabrication and Repair of Cartilage Defects with a Novel Acellular Cartilage Matrix Scaffold

The objectives of this study were to develop a three-dimensional acellular cartilage matrix (ACM) and investigate its possibility for use as a scaffold in cartilage tissue engineering. Bovine articular cartilage was decellularized sequentially with trypsin, nuclease solution, hypotonic buffer, and Triton x 100 solution; molded with freeze-drying process; and cross-linked by ultraviolet irradiation. Histological and biochemical analysis showed that the ACM was devoid of cells and still maintained the collagen and glycosaminoglycan components of cartilage. Scanning electronic microscopy and mercury intrusion porosimetry showed that the ACM had a sponge-like structure of high porosity. The ACM scaffold had good biocompatibility with cultured rabbit bone marrow mesenchymal stem cells with no indication of cytotoxicity both in contact and in extraction assays. The cartilage defects repair in rabbit knees with the mesenchymal stem cell-ACM constructs had a significant improvement of histological scores when compared to the control groups at 6 and 12 weeks. In summary, the ACM possessed the characteristics that afford it as a potential scaffold for cartilage tissue engineering.