Gaussian processes for classification: mean field algorithms:mean-field algorithms

We derive a mean field algorithm for binary classification with Gaussian processes which is based on the TAP approach originally proposed in Statistical Physics of disordered systems. The theory also yields an approximate leave-one-out estimator for the generalization error which is computed with no extra computational cost. We show that from the TAP approach, it is possible to derive both a simpler 'naive' mean field theory and support vector machines (SVM) as limiting cases. For both mean field algorithms and support vectors machines, simulation results for three small benchmark data sets are presented. They show 1. that one may get state of the art performance by using the leave-one-out estimator for model selection and 2. the built-in leave-one-out estimators are extremely precise when compared to the exact leave-one-out estimate. The latter result is a taken as a strong support for the internal consistency of the mean field approach.

Gaussian processes and SVM: Mean field results and leave-one-out

In this chapter, we elaborate on the well-known relationship between Gaussian processes (GP) and Support Vector Machines (SVM). Secondly, we present approximate solutions for two computational problems arising in GP and SVM. The first one is the calculation of the posterior mean for GP classifiers using a `naive' mean field approach. The second one is a leave-one-out estimator for the generalization error of SVM based on a linear response method. Simulation results on a benchmark dataset show similar performances for the GP mean field algorithm and the SVM algorithm. The approximate leave-one-out estimator is found to be in very good agreement with the exact leave-one-out error.

Evaluation of an open-field autorefractor's ability to measure refraction and hence potential to assess objective accommodation in pseudophakes

Background: To evaluate the accuracy of an open-field autorefractor compared with subjective refraction in pseudophakes and hence its ability to assess objective eye focus with intraocular lenses (IOLs). Methods: Objective refraction was measured at 6 m using the Shin-Nippon NVision-K 5001/Grand Seiko WR-5100K open-field autorefractor (five repeats) and by subjective refraction on 141 eyes implanted with a spherical (Softec1 n=53), aspherical (SoftecHD n=37) or accommodating (1CU n=22; Tetraflex n=29) IOL. Autorefraction was repeated 2 months later. Results: The autorefractor prescription was similar (average difference: 0.09±0.53 D; p=0.19) to that found by subjective refraction, with ~71% within ±0.50 D. The horizontal cylindrical components were similar (difference: 0.00±0.39 D; p=0.96), although the oblique (J45) autorefractor cylindrical vector was slightly more negative (by -0.06±0.25 D; p=0.06) than the subjective refraction. The results were similar for each of the IOL designs except for the spherical IOL, where the mean spherical equivalent difference between autorefraction and subjective was more hypermetropic than the Tetraflex accommodating IOL (F=2.77, p=0.04). The intrasession repeatability was

Bayesian inference for wind field retrieval

In many problems in spatial statistics it is necessary to infer a global problem solution by combining local models. A principled approach to this problem is to develop a global probabilistic model for the relationships between local variables and to use this as the prior in a Bayesian inference procedure. We show how a Gaussian process with hyper-parameters estimated from Numerical Weather Prediction Models yields meteorologically convincing wind fields. We use neural networks to make local estimates of wind vector probabilities. The resulting inference problem cannot be solved analytically, but Markov Chain Monte Carlo methods allow us to retrieve accurate wind fields.

A pragmatic Bayesian approach to wind field retrieval

The ERS-1 Satellite was launched in July 1991 by the European Space Agency into a polar orbit at about 800 km, carrying a C-band scatterometer. A scatterometer measures the amount of backscatter microwave radiation reflected by small ripples on the ocean surface induced by sea-surface winds, and so provides instantaneous snap-shots of wind flow over large areas of the ocean surface, known as wind fields. Inherent in the physics of the observation process is an ambiguity in wind direction; the scatterometer cannot distinguish if the wind is blowing toward or away from the sensor device. This ambiguity implies that there is a one-to-many mapping between scatterometer data and wind direction. Current operational methods for wind field retrieval are based on the retrieval of wind vectors from satellite scatterometer data, followed by a disambiguation and filtering process that is reliant on numerical weather prediction models. The wind vectors are retrieved by the local inversion of a forward model, mapping scatterometer observations to wind vectors, and minimising a cost function in scatterometer measurement space. This thesis applies a pragmatic Bayesian solution to the problem. The likelihood is a combination of conditional probability distributions for the local wind vectors given the scatterometer data. The prior distribution is a vector Gaussian process that provides the geophysical consistency for the wind field. The wind vectors are retrieved directly from the scatterometer data by using mixture density networks, a principled method to model multi-modal conditional probability density functions. The complexity of the mapping and the structure of the conditional probability density function are investigated. A hybrid mixture density network, that incorporates the knowledge that the conditional probability distribution of the observation process is predominantly bi-modal, is developed. The optimal model, which generalises across a swathe of scatterometer readings, is better on key performance measures than the current operational model. Wind field retrieval is approached from three perspectives. The first is a non-autonomous method that confirms the validity of the model by retrieving the correct wind field 99% of the time from a test set of 575 wind fields. The second technique takes the maximum a posteriori probability wind field retrieved from the posterior distribution as the prediction. For the third technique, Markov Chain Monte Carlo (MCMC) techniques were employed to estimate the mass associated with significant modes of the posterior distribution, and make predictions based on the mode with the greatest mass associated with it. General methods for sampling from multi-modal distributions were benchmarked against a specific MCMC transition kernel designed for this problem. It was shown that the general methods were unsuitable for this application due to computational expense. On a test set of 100 wind fields the MAP estimate correctly retrieved 72 wind fields, whilst the sampling method correctly retrieved 73 wind fields.

Gravity field recovery using two low satellites in different orbital planes

This study is concerned with gravity field recovery from low-low satellite to satellite range rate data. An improvement over a coplanar mission is predicted in the errors associated with certain parts of the geopotential by the separation of the orbital planes of the two satellites. Using Hill's equations an analytical scheme to model the range rate residuals is developed. It is flexible enough to model equally well the residuals between pairs of satellites in the same orbital plane or whose planes are separated in right ascension. The possible benefits of such an orientation to gravity field recovery from range rate data can therefore be analysed, and this is done by means of an extensive error analysis. The results of this analysis show that for an optimal planar mission improvements can be made by separating the satellites in right ascension. Gravity field recoveries are performed in order to verify and gauge the limitations of the analytical model, and to support the results of the error analysis. Finally the possible problem of the differential decay rates of two satellites due to the diurnal bulge are evaluated.

Design and validity of a miniaturized open-field aberrometer

PURPOSE: To validate a new miniaturised, open-field wavefront device which has been developed with the capacity to be attached to an ophthalmic surgical microscope or slit-lamp. SETTING: Solihull Hospital and Aston University, Birmingham, UK DESIGN: Comparative non-interventional study. METHODS: The dynamic range of the Aston Aberrometer was assessed using a calibrated model eye. The validity of the Aston Aberrometer was compared to a conventional desk mounted Shack-Hartmann aberrometer (Topcon KR1W) by measuring the refractive error and higher order aberrations of 75 dilated eyes with both instruments in random order. The Aston Aberrometer measurements were repeated five times to assess intra-session repeatability. Data was converted to vector form for analysis. RESULTS: The Aston Aberrometer had a large dynamic range of at least +21.0 D to -25.0 D. It gave similar measurements to a conventional aberrometer for mean spherical equivalent (mean difference ± 95% confidence interval: 0.02 ± 0.49D; correlation: r=0.995, p<0.001), astigmatic components (J0: 0.02 ± 0.15D; r=0.977, p<0.001; J45: 0.03 ± 0.28; r=0.666, p<0.001) and higher order aberrations RMS (0.02 ± 0.20D; r=0.620, p<0.001). Intraclass correlation coefficient assessments of intra-sessional repeatability for the Aston Aberrometer were excellent (spherical equivalent =1.000, p<0.001; astigmatic components J0 =0.998, p<0.001, J45=0.980, p<0.01; higher order aberrations RMS =0.961, p<0.001). CONCLUSIONS: The Aston Aberrometer gives valid and repeatable measures of refractive error and higher order aberrations over a large range. As it is able to measure continuously, it can provide direct feedback to surgeons during intraocular lens implantations and corneal surgery as to the optical status of the visual system.

Determining planar multiple sound-soft obstacles from scattered acoustic fields

An inverse problem is considered where the structure of multiple sound-soft planar obstacles is to be determined given the direction of the incoming acoustic field and knowledge of the corresponding total field on a curve located outside the obstacles. A local uniqueness result is given for this inverse problem suggesting that the reconstruction can be achieved by a single incident wave. A numerical procedure based on the concept of the topological derivative of an associated cost functional is used to produce images of the obstacles. No a priori assumption about the number of obstacles present is needed. Numerical results are included showing that accurate reconstructions can be obtained and that the proposed method is capable of finding both the shapes and the number of obstacles with one or a few incident waves.

Recovering boundary data in planar heat conduction using boundary integral equation method

We consider a Cauchy problem for the heat equation, where the temperature field is to be reconstructed from the temperature and heat flux given on a part of the boundary of the solution domain. We employ a Landweber type method proposed in [2], where a sequence of mixed well-posed problems are solved at each iteration step to obtain a stable approximation to the original Cauchy problem. We develop an efficient boundary integral equation method for the numerical solution of these mixed problems, based on the method of Rothe. Numerical examples are presented both with exact and noisy data, showing the efficiency and stability of the proposed procedure and approximations.

Efficient Computing of some Vector Operations over GF(3) and GF(4)

The problem of efficient computing of the affine vector operations (addition of two vectors and multiplication of a vector by a scalar over GF (q)), and also the weight of a given vector, is important for many problems in coding theory, cryptography, VLSI technology etc. In this paper we propose a new way of representing vectors over GF (3) and GF (4) and we describe an efficient performance of these affine operations. Computing weights of binary vectors is also discussed.

Fast and slowly evolving vector solitons in mode-locked fibre lasers

We report on a new vector model of an erbium-doped fibre laser mode locked with carbon nanotubes. This model goes beyond the limitations of the previously used models based on either coupled nonlinear Schrödinger or Ginzburg-Landau equations. Unlike the previous models, it accounts for the vector nature of the interaction between an optical field and an erbium-doped active medium, slow relaxation dynamics of erbium ions, linear birefringence in a fibre, linear and circular birefringence of a laser cavity caused by in-cavity polarization controller and light-induced anisotropy caused by elliptically polarized pump field. Interplay of aforementioned factors changes coherent coupling of two polarization modes at a long time scale and so results in a new family of vector solitons (VSs) with fast and slowly evolving states of polarization. The observed VSs can be of interest in secure communications, trapping and manipulation of atoms and nanoparticles, control of magnetization in data storage devices and many other areas.

Assembly accuracy analysis for small components with a planar surface in large-scale metrology

Large-scale mechanical products, such as aircraft and rockets, consist of large numbers of small components, which introduce additional difficulty for assembly accuracy and error estimation. Planar surfaces as key product characteristics are usually utilised for positioning small components in the assembly process. This paper focuses on assembly accuracy analysis of small components with planar surfaces in large-scale volume products. To evaluate the accuracy of the assembly system, an error propagation model for measurement error and fixture error is proposed, based on the assumption that all errors are normally distributed. In this model, the general coordinate vector is adopted to represent the position of the components. The error transmission functions are simplified into a linear model, and the coordinates of the reference points are composed by theoretical value and random error. The installation of a Head-Up Display is taken as an example to analyse the assembly error of small components based on the propagation model. The result shows that the final coordination accuracy is mainly determined by measurement error of the planar surface in small components. To reduce the uncertainty of the plane measurement, an evaluation index of measurement strategy is presented. This index reflects the distribution of the sampling point set and can be calculated by an inertia moment matrix. Finally, a practical application is introduced for validating the evaluation index.

Identifying DNA-binding proteins by combining support vector machine and PSSM distance transformation

Background: DNA-binding proteins play a pivotal role in various intra- and extra-cellular activities ranging from DNA replication to gene expression control. Identification of DNA-binding proteins is one of the major challenges in the field of genome annotation. There have been several computational methods proposed in the literature to deal with the DNA-binding protein identification. However, most of them can't provide an invaluable knowledge base for our understanding of DNA-protein interactions. Results: We firstly presented a new protein sequence encoding method called PSSM Distance Transformation, and then constructed a DNA-binding protein identification method (SVM-PSSM-DT) by combining PSSM Distance Transformation with support vector machine (SVM). First, the PSSM profiles are generated by using the PSI-BLAST program to search the non-redundant (NR) database. Next, the PSSM profiles are transformed into uniform numeric representations appropriately by distance transformation scheme. Lastly, the resulting uniform numeric representations are inputted into a SVM classifier for prediction. Thus whether a sequence can bind to DNA or not can be determined. In benchmark test on 525 DNA-binding and 550 non DNA-binding proteins using jackknife validation, the present model achieved an ACC of 79.96%, MCC of 0.622 and AUC of 86.50%. This performance is considerably better than most of the existing state-of-the-art predictive methods. When tested on a recently constructed independent dataset PDB186, SVM-PSSM-DT also achieved the best performance with ACC of 80.00%, MCC of 0.647 and AUC of 87.40%, and outperformed some existing state-of-the-art methods. Conclusions: The experiment results demonstrate that PSSM Distance Transformation is an available protein sequence encoding method and SVM-PSSM-DT is a useful tool for identifying the DNA-binding proteins. A user-friendly web-server of SVM-PSSM-DT was constructed, which is freely accessible to the public at the web-site on http://bioinformatics.hitsz.edu.cn/PSSM-DT/.

Field programmable gate array based target detection and gesture recognition

The move from Standard Definition (SD) to High Definition (HD) represents a six times increases in data, which needs to be processed. With expanding resolutions and evolving compression, there is a need for high performance with flexible architectures to allow for quick upgrade ability. The technology advances in image display resolutions, advanced compression techniques, and video intelligence. Software implementation of these systems can attain accuracy with tradeoffs among processing performance (to achieve specified frame rates, working on large image data sets), power and cost constraints. There is a need for new architectures to be in pace with the fast innovations in video and imaging. It contains dedicated hardware implementation of the pixel and frame rate processes on Field Programmable Gate Array (FPGA) to achieve the real-time performance. ^ The following outlines the contributions of the dissertation. (1) We develop a target detection system by applying a novel running average mean threshold (RAMT) approach to globalize the threshold required for background subtraction. This approach adapts the threshold automatically to different environments (indoor and outdoor) and different targets (humans and vehicles). For low power consumption and better performance, we design the complete system on FPGA. (2) We introduce a safe distance factor and develop an algorithm for occlusion occurrence detection during target tracking. A novel mean-threshold is calculated by motion-position analysis. (3) A new strategy for gesture recognition is developed using Combinational Neural Networks (CNN) based on a tree structure. Analysis of the method is done on American Sign Language (ASL) gestures. We introduce novel point of interests approach to reduce the feature vector size and gradient threshold approach for accurate classification. (4) We design a gesture recognition system using a hardware/ software co-simulation neural network for high speed and low memory storage requirements provided by the FPGA. We develop an innovative maximum distant algorithm which uses only 0.39% of the image as the feature vector to train and test the system design. Database set gestures involved in different applications may vary. Therefore, it is highly essential to keep the feature vector as low as possible while maintaining the same accuracy and performance^

Simplified PRBMs of spatial compliant multi-beam modules for planar motion

PRBMs (pseudo-rigid-body models) have been becoming important engineering technologies/methods in the field of compliant mechanisms to simplify the design and analysis through the use of the knowledge body of rigid-body mechanisms coupling with springs. This article addresses the PRBMs of spatial multi-beam modules for planar motion, which are composed of three or more symmetrical wire/slender beams parallel to each other where the planar twisting DOF (degree of freedom) is assumed to be very small for specific applications/loading conditions. Simplified PRBMs are firstly proposed through replacing each beam in spatial multi-beam module with a rigid-body link plus two identical spherical joints at its two ends. The characteristics factor, bending stiffness and twisting stiffness for the spherical joint are determined. Load-displacement equations are then derived for a class of spatial multi-beam modules and general spatial multi-beam modules using the virtual work principle and kinematic relationships. Finally, nonlinear FEA (finite element analysis) is employed with comparisons with the PRBMs. The present PRBMs have shown the ability to predict the primary nonlinear constraint characteristics such as load-stiffening effect, cross-axis coupling in the two primary translational directions and buckling load.