Beyond dataset bias: Multi-task unaligned shared knowledge transfer

Many visual datasets are traditionally used to analyze the performance of different learning techniques. The evaluation is usually done within each dataset, therefore it is questionable if such results are a reliable indicator of true generalization ability. We propose here an algorithm to exploit the existing data resources when learning on a new multiclass problem. Our main idea is to identify an image representation that decomposes orthogonally into two subspaces: a part specific to each dataset, and a part generic to, and therefore shared between, all the considered source sets. This allows us to use the generic representation as un-biased reference knowledge for a novel classification task. By casting the method in the multi-view setting, we also make it possible to use different features for different databases. We call the algorithm MUST, Multitask Unaligned Shared knowledge Transfer. Through extensive experiments on five public datasets, we show that MUST consistently improves the cross-datasets generalization performance. © 2013 Springer-Verlag.

Mass transfer resistance in the production of high impact polypropylene

Mass transfer resistance in the production of high impact polypropylene (hiPP) produced by a two-stage slurry/gas polymerization was investigated by field-emission scanning electron microscopy (SEM) and transmission electron microscopy (TEM) techniques. It is found that the formation of ethylene-propylene copolymer (EPR) phases in polypropylene (iPP) particle produced in the first stage slurry polymerization exhibits a developing process from exterior to interior

Study of cupric hexacyanoferrate-modified platinum electrodes using probe beam deflection and electrochemical quartz crystal microbalance techniques

The ion exchange mechanism accompanying the oxidation/reduction processes of cupric hexacyanoferrate-modified platinum electrodes in different aqueous electrolyte solutions has been studied by means of in situ probe beam deflection and the electrochemical quartz crystal microbalance technique. The results demonstrate that the charge neutrality of the film during the reoxidation/reduction process is accomplished predominantly by the movement of cations, but anions and/or solvent are also participator(s). Moreover, in KHC8H4O4 (potassium biphthalate) solution, the EQCM data obtained from chronoamperometry experiment are more complicated than those in KCl and K2SO4 solutions. (C) 1997 Elsevier Science Ltd.

APPLICATIONS OF RAPID SCAN SPECTROELECTROCHEMISTRY - DETERMINATION OF HETEROGENEOUS ELECTRON-TRANSFER RATE-CONSTANT OF BILIVERDIN OXIDATION

The rapid scan spectrometer was used to determine the heterogeneous electron transfer rate parameters for the oxidation of Biliverdin in DMF by single potential step thin layer spectroelectrochemical techniques and yielded an average formal heterogeneous electron transfer rate constant K(s, h)0' = 2.45 (+/-0.12) x 10(-4) cm s-1, electrochemical transfer coefficient alpha = 0.694+/-0.008. The oxidation process of Biliverdin was also studied and the formal potential E0 = 0.637 V (vs. Ag/AgCl) was obtained.

Techniques for engineering quantum states

In this thesis I theoretically study quantum states of ultracold atoms. The majority of the Chapters focus on engineering specific quantum states of single atoms with high fidelity in experimentally realistic systems. In the sixth Chapter, I investigate the stability and dynamics of new multidimensional solitonic states that can be created in inhomogeneous atomic Bose-Einstein condensates. In Chapter three I present two papers in which I demonstrate how the coherent tunnelling by adiabatic passage (CTAP) process can be implemented in an experimentally realistic atom chip system, to coherently transfer the centre-of-mass of a single atom between two spatially distinct magnetic waveguides. In these works I also utilise GPU (Graphics Processing Unit) computing which offers a significant performance increase in the numerical simulation of the Schrödinger equation. In Chapter four I investigate the CTAP process for a linear arrangement of radio frequency traps where the centre-of-mass of both, single atoms and clouds of interacting atoms, can be coherently controlled. In Chapter five I present a theoretical study of adiabatic radio frequency potentials where I use Floquet theory to more accurately model situations where frequencies are close and/or field amplitudes are large. I also show how one can create highly versatile 2D adiabatic radio frequency potentials using multiple radio frequency fields with arbitrary field orientation and demonstrate their utility by simulating the creation of ring vortex solitons. In the sixth Chapter I discuss the stability and dynamics of a family of multidimensional solitonic states created in harmonically confined Bose-Einstein condensates. I demonstrate that these solitonic states have interesting dynamical instabilities, where a continuous collapse and revival of the initial state occurs. Through Bogoliubov analysis, I determine the modes responsible for the observed instabilities of each solitonic state and also extract information related to the time at which instability can be observed.

Restoration of beta-adrenergic signaling in failing cardiac ventricular myocytes via adenoviral-mediated gene transfer.

Cardiovascular gene therapy is a novel approach to the treatment of diseases such as congestive heart failure (CHF). Gene transfer to the heart would allow for the replacement of defective or missing cellular proteins that may improve cardiac performance. Our laboratory has been focusing on the feasibility of restoring beta-adrenergic signaling deficiencies that are a characteristic of chronic CHF. We have now studied isolated ventricular myocytes from rabbits that have been chronically paced to produce hemodynamic failure. We document molecular beta-adrenergic signaling defects including down-regulation of myocardial beta-adrenergic receptors (beta-ARs), functional beta-AR uncoupling, and an up-regulation of the beta-AR kinase (betaARK1). Adenoviral-mediated gene transfer of the human beta2-AR or an inhibitor of betaARK1 to these failing myocytes led to the restoration of beta-AR signaling. These results demonstrate that defects present in this critical myocardial signaling pathway can be corrected in vitro using genetic modification and raise the possibility of novel inotropic therapies for CHF including the inhibition of betaARK1 activity in the heart.

Potentiation of beta-adrenergic signaling by adenoviral-mediated gene transfer in adult rabbit ventricular myocytes.

Our laboratory has been testing the hypothesis that genetic modulation of the beta-adrenergic signaling cascade can enhance cardiac function. We have previously shown that transgenic mice with cardiac overexpression of either the human beta2-adrenergic receptor (beta2AR) or an inhibitor of the beta-adrenergic receptor kinase (betaARK), an enzyme that phosphorylates and uncouples agonist-bound receptors, have increased myocardial inotropy. We now have created recombinant adenoviruses encoding either the beta2AR (Adeno-beta2AR) or a peptide betaARK inhibitor (consisting of the carboxyl terminus of betaARK1, Adeno-betaARKct) and tested their ability to potentiate beta-adrenergic signaling in cultured adult rabbit ventricular myocytes. As assessed by radioligand binding, Adeno-beta2AR infection led to approximately 20-fold overexpression of beta-adrenergic receptors. Protein immunoblots demonstrated the presence of the Adeno-betaARKct transgene. Both transgenes significantly increased isoproterenol-stimulated cAMP as compared to myocytes infected with an adenovirus encoding beta-galactosidase (Adeno-betaGal) but did not affect the sarcolemmal adenylyl cyclase response to Forskolin or NaF. beta-Adrenergic agonist-induced desensitization was significantly inhibited in Adeno-betaARKct-infected myocytes (16+/-2%) as compared to Adeno-betaGal-infected myocytes (37+/-1%, P < 0.001). We conclude that recombinant adenoviral gene transfer of the beta2AR or an inhibitor of betaARK-mediated desensitization can potentiate beta-adrenergic signaling.

Impacts of Mountaintop Removal Coal Mining on the Mud River, West Virginia: Selenium Accumulation, Trophic Transfer, and Toxicity in Fish

Selenium (Se) is a micronutrient necessary for the function of a variety of important enzymes; Se also exhibits a narrow range in concentrations between essentiality and toxicity. Oviparous vertebrates such as birds and fish are especially sensitive to Se toxicity, which causes reproductive impairment and defects in embryo development. Selenium occurs naturally in the Earth's crust, but it can be mobilized by a variety of anthropogenic activities, including agricultural practices, coal burning, and mining.

Mountaintop removal/valley fill (MTR/VF) coal mining is a form of surface mining found throughout central Appalachia in the United States that involves blasting off the tops of mountains to access underlying coal seams. Spoil rock from the mountain is placed into adjacent valleys, forming valley fills, which bury stream headwaters and negatively impact surface water quality. This research focused on the biological impacts of Se leached from MTR/VF coal mining operations located around the Mud River, West Virginia.

In order to assess the status of Se in a lotic (flowing) system such as the Mud River, surface water, insects, and fish samples including creek chub (Semotilus atromaculatus) and green sunfish (Lepomis cyanellus) were collected from a mining impacted site as well as from a reference site not impacted by mining. Analysis of samples from the mined site showed increased conductivity and Se in the surface waters compared to the reference site in addition to increased concentrations of Se in insects and fish. Histological analysis of mined site fish gills showed a lack of normal parasites, suggesting parasite populations may be disrupted due to poor water quality. X-ray absorption near edge spectroscopy techniques were used to determine the speciation of Se in insect and creek chub samples. Insects contained approximately 40-50% inorganic Se (selenate and selenite) and 50-60% organic Se (Se-methionine and Se-cystine) while fish tissues contained lower proportions of inorganic Se than insects, instead having higher proportions of organic Se in the forms of methyl-Se-cysteine, Se-cystine, and Se-methionine.

Otoliths, calcified inner ear structures, were also collected from Mud River creek chubs and green sunfish and analyzed for Se content using laser ablation inductively couple mass spectrometry (LA-ICP-MS). Significant differences were found between the two species of fish, based on the concentrations of otolith Se. Green sunfish otoliths from all sites contained background or low concentrations of otolith Se (< 1 µg/g) that were not significantly different between mined and unmined sites. In contrast creek chub otoliths from the historically mined site contained much higher (≥ 5 µg/g, up to approximately 68 µg/g) concentrations of Se than for the same species in the unmined site or for the green sunfish. Otolith Se concentrations were related to muscle Se concentrations for creek chubs (R2 = 0.54, p = 0.0002 for the last 20% of the otolith Se versus muscle Se) while no relationship was observed for green sunfish.

Additional experiments using biofilms grown in the Mud River showed increased Se in mined site biofilms compared to the reference site. When we fed fathead minnows (Pimephales promelas) on these biofilms in the laboratory they accumulated higher concentrations of Se in liver and ovary tissues compared to fathead minnows fed on reference site biofilms. No differences in Se accumulation were found in muscle from either treatment group. Biofilms were also centrifuged and separated into filamentous green algae and the remaining diatom fraction. The majority of Se was found in the diatom fraction with only about 1/3rd of total biofilm Se concentration present in the filamentous green algae fraction

Finally, zebrafish (Danio rerio) embryos were exposed to aqueous Se in the form of selenate, selenite, and L-selenomethionine in an attempt to determine if oxidative stress plays a role in selenium embryo toxicity. Selenate and selenite exposure did not induce embryo deformities (lordosis and craniofacial malformation). L-selenomethionine, however, induced significantly higher deformity rates at 100 µg/L compared to controls. Antioxidant rescue of L-selenomethionime induced deformities was attempted in embryos using N-acetylcysteine (NAC). Pretreatment with NAC significantly reduced deformities in the zebrafish embryos secondarily treated with L-selenomethionine, suggesting that oxidative stress may play a role in Se toxicity. Selenite exposure also induced a 6.6-fold increase in glutathione-S-transferase pi class 2 gene expression, which is involved in xenobiotic transformation. No changes in gene expression were observed for selenate or L-selenomethionine-exposed embryos.

The findings in this dissertation contribute to the understanding of how Se bioaccumulates in a lotic system and is transferred through a simulated foodweb in addition to further exploring oxidative stress as a potential mechanism for Se-induced embryo toxicity. Future studies should continue to pursue the role of oxidative stress and other mechanisms in Se toxicity and the biotransformation of Se in aquatic ecosystems.

The development and validation of computational MHD techniques for the modelling of metal production processes

Magnetic fields are used in a number of processes related to the extraction of metals, production of alloys and the shaping of metal components. Computational techniques have an increasingly important role to play in the simulation of such processes, since it is often difficult or very costly to conduct experiments in the high temperature conditions encountered and the complex interaction of fluid flow, heat transfer and magnetic fields means simple analytic models are often far removed from reality. In this paper an overview of the computational activity at the University of Greenwich is given in this area, covering the past ten years. The overview is given from the point of view of the modeller and within the space limitations imposed by the format it covers the numerical methods used, attempts at validation against experiments or analytic procedures; it highlights successes, but also some failures. A broad range of models is covered in the review (and accompanying lecture), used to simulate (a) A-C field applications: induction melting, magnetic confinement and levitation, casting and (b) D-C field applications such as: arc welding and aluminium electroloysis. Most of these processes involve phase change of the metal (melting or solidification), the presence of a dynamic free surface and turbulent flow. These issues affect accuracy and need to be address by the modeller.

Computational modeling techniques for reliability of electronic components on printed circuit boards

This paper describes modeling technology and its use in providing data governing the assembly and subsequent reliability of electronic chip components on printed circuit boards (PCBs). Products, such as mobile phones, camcorders, intelligent displays, etc., are changing at a tremendous rate where newer technologies are being applied to satisfy the demands for smaller products with increased functionality. At ever decreasing dimensions, and increasing number of input/output connections, the design of these components, in terms of dimensions and materials used, is playing a key role in determining the reliability of the final assembly. Multiphysics modeling techniques are being adopted to predict a range of interacting physics-based phenomena associated with the manufacturing process. For example, heat transfer, solidification, marangoni fluid flow, void movement, and thermal-stress. The modeling techniques used are based on finite volume methods that are conservative and take advantage of being able to represent the physical domain using an unstructured mesh. These techniques are also used to provide data on thermal induced fatigue which is then mapped into product lifetime predictions.

Quantification of oxygenated volatile organic compounds in seawater by membrane inlet-proton transfer reaction/mass spectrometry

The role of the ocean in the cycling of oxygenated volatile organic compounds (OVOCs) remains largely unanswered due to a paucity of datasets. We describe the method development of a membrane inlet-proton transfer reaction/mass spectrometer (MI-PTR/MS) as an efficient method of analysing methanol, acetaldehyde and acetone in seawater. Validation of the technique with water standards shows that the optimised responses are linear and reproducible. Limits of detection are 27 nM for methanol, 0.7 nM for acetaldehyde and 0.3 nM for acetone. Acetone and acetaldehyde concentrations generated by MI-PTR/MS are compared to a second, independent method based on purge and trap-gas chromatography/flame ionisation detection (P&T-GC/FID) and show excellent agreement. Chromatographic separation of isomeric species acetone and propanal permits correction to mass 59 signal generated by the PTR/MS and overcomes a known uncertainty in reporting acetone concentrations via mass spectrometry. A third bioassay technique using radiolabelled acetone further supported the result generated by this method. We present the development and optimisation of the MI-PTR/MS technique as a reliable and convenient tool for analysing seawater samples for these trace gases. We compare this method with other analytical techniques and discuss its potential use in improving the current understanding of the cycling of oceanic OVOCs.

Transfer Across the Air-Sea Interface

The efficiency of transfer of gases and particles across the air-sea interface is controlled by several physical, biological and chemical processes in the atmosphere and water which are described here (including waves, large- and small-scale turbulence, bubbles, sea spray, rain and surface films). For a deeper understanding of relevant transport mechanisms, several models have been developed, ranging from conceptual models to numerical models. Most frequently the transfer is described by various functional dependencies of the wind speed, but more detailed descriptions need additional information. The study of gas transfer mechanisms uses a variety of experimental methods ranging from laboratory studies to carbon budgets, mass balance methods, micrometeorological techniques and thermographic techniques. Different methods resolve the transfer at different scales of time and space; this is important to take into account when comparing different results. Air-sea transfer is relevant in a wide range of applications, for example, local and regional fluxes, global models, remote sensing and computations of global inventories. The sensitivity of global models to the description of transfer velocity is limited; it is however likely that the formulations are more important when the resolution increases and other processes in models are improved. For global flux estimates using inventories or remote sensing products the accuracy of the transfer formulation as well as the accuracy of the wind field is crucial.

The transfer of energy between electrons and ions in solids

In this review we consider those processes in condensed matter that involve the irreversible flow of energy between electrons and nuclei that follows from a system being taken out of equilibrium. We survey some of the more important experimental phenomena associated with these processes, followed by a number of theoretical techniques for studying them. The techniques considered are those that can be applied to systems containing many nonequivalent atoms. They include both perturbative approaches (Fermi's Golden Rule and non-equilibrium Green's functions) and molecular dynamics based (the Ehrenfest approximation, surface hopping, semi-classical Gaussian wavefunction methods and correlated electron-ion dynamics). These methods are described and characterized, with indications of their relative merits.

Measuring the solubility of benzoic acid in room temperature ionic liquids using chronoamperometric techniques

The electrochemical reduction of benzoic acid (BZA) has been studied at platinum micro-electrodes (10 and 2 mu m diameters) in acetonitrile (MeCN) and six room temperature ionic liquids (RTILs): [C(2)mim][NTf2], [C(4)min][NTf2], [C(4)mpyrr][NTf2], [C(4)mim][BF4], [C(4)mim][NO3] and [C(4)mim][PF6] (where [C(n)mim](+)=1-alkyl-3-methylimidazolium, [NTf2](-)=bis(trifluoromethylsulphonyl)imide, [C(4)mpyrr](+)=N-butyl-N-methylpyrrolidinium, [BF4](-)=tetrafluoroborate, [NO3](-)=nitrate and [PF6] = hexafluorophosphate). Based on the theoretical fitting to experimental chronoamperometric transients in [C4mpyrr][NTf2] and MeCN at several concentrations and on different size electrodes, it is suggested that a fast chemical step preceeds the electron transfer step in a CE mechanism (given below) in both RTILs and MeCN, leading to the appearance of a simple one-electron transfer mechanism.

Simulation of coupled heat and moisture transfer in air-conditioned buildings

The simultaneous heat and moisture transfer in the building envelope has an important influence on the indoor environment and the overall performance of buildings. In this paper, a model for predicting whole building heat and moisture transfer was presented. Both heat and moisture transfer in the building envelope and indoor air were simultaneously considered; their interactions were modeled. The coupled model takes into account most of the main hygrothermal effects in buildings. The coupled system model was implemented in MATLAB-Simulink, and validated by using a series of published testing tools. The new program was applied to investigate the moisture transfer effect on indoor air humidity and building energy consumption under different climates. The results show that the use of more detailed simulation routines can result in improvements to the building's design for energy optimisation through the choice of proper hygroscopic materials, which would not be indicated by simpler calculation techniques.