Variational inference for diffusion processes

Diffusion processes are a family of continuous-time continuous-state stochastic processes that are in general only partially observed. The joint estimation of the forcing parameters and the system noise (volatility) in these dynamical systems is a crucial, but non-trivial task, especially when the system is nonlinear and multimodal. We propose a variational treatment of diffusion processes, which allows us to compute type II maximum likelihood estimates of the parameters by simple gradient techniques and which is computationally less demanding than most MCMC approaches. We also show how a cheap estimate of the posterior over the parameters can be constructed based on the variational free energy.

The development of a solar wall module using a selective surface

A need was indicated for the identification of a possible new solar energy product to improve the sales potential of a metal film with a selective surface, manufactured by the industriaI sponsor of this project (INCO). A possible way of overcoming the disadvantageous economics of solar energy collection was identified. This utilised the collection of solar energy by the walls of buildings constructed in such a manner as to allow the transfer of energy into the building, whilst providing adequate thermal insulation in the absence of sunlight. The actual collection element of the wall, being metallic, is also capable of performing the function of a low temperature heating .system in the absence of sunlight. As a result of this, the proposed system, by displacing both the wall and centraI heating system which would otherwise be necessary, demonstrates economic benefits over systems which are constructed solely for the purpose of collecting solar energy. The necessary thermodynamic and meteorological. characteristics and data: are established, and applied to a typical urban site in the North of England, for a typical average year, with and without a shading device incorporated into the construction. It is concluded that the proposed system may offer considerable benefit in reducing the effective heating season in all orientations of wall.

Control integrated finite state machine design for photovoltaic module integrated converter

The operation state of photovoltaic Module Integrated Converter (MIC) is subjected to change due to different source and load conditions, while state-swap is usually implemented with flow chart based sequential controller in the past research. In this paper, the signatures for different operational states are evaluated and investigated, which lead to an effective control integrated finite state machine (CIFSM), providing real-time state-swap as fast as the local control loop. The proposed CIFSM is implemented digitally for a boost type MIC prototype and tested under a variety of load and source conditions. The test results prove the effectiveness of the proposed CIFSM design.

Replication-based inference algorithms for hard computational problems

Inference algorithms based on evolving interactions between replicated solutions are introduced and analyzed on a prototypical NP-hard problem: the capacity of the binary Ising perceptron. The efficiency of the algorithm is examined numerically against that of the parallel tempering algorithm, showing improved performance in terms of the results obtained, computing requirements and simplicity of implementation. © 2013 American Physical Society.

Estimation and inference under economic restrictions

Estimation of economic relationships often requires imposition of constraints such as positivity or monotonicity on each observation. Methods to impose such constraints, however, vary depending upon the estimation technique employed. We describe a general methodology to impose (observation-specific) constraints for the class of linear regression estimators using a method known as constraint weighted bootstrapping. While this method has received attention in the nonparametric regression literature, we show how it can be applied for both parametric and nonparametric estimators. A benefit of this method is that imposing numerous constraints simultaneously can be performed seamlessly. We apply this method to Norwegian dairy farm data to estimate both unconstrained and constrained parametric and nonparametric models.

Message-passing for inference and optimization of real variables on sparse graphs

The inference and optimization in sparse graphs with real variables is studied using methods of statistical mechanics. Efficient distributed algorithms for the resource allocation problem are devised. Numerical simulations show excellent performance and full agreement with the theoretical results. © Springer-Verlag Berlin Heidelberg 2006.

Variational mean-field algorithm for efficient inference in large systems of stochastic differential equations

This work introduces a Gaussian variational mean-field approximation for inference in dynamical systems which can be modeled by ordinary stochastic differential equations. This new approach allows one to express the variational free energy as a functional of the marginal moments of the approximating Gaussian process. A restriction of the moment equations to piecewise polynomial functions, over time, dramatically reduces the complexity of approximate inference for stochastic differential equation models and makes it comparable to that of discrete time hidden Markov models. The algorithm is demonstrated on state and parameter estimation for nonlinear problems with up to 1000 dimensional state vectors and compares the results empirically with various well-known inference methodologies.

Inference of helicopter airframe condition

The goal of this paper is to model normal airframe conditions for helicopters in order to detect changes. This is done by inferring the flying state using a selection of sensors and frequency bands that are best for discriminating between different states. We used non-linear state-space models (NLSSM) for modelling flight conditions based on short-time frequency analysis of the vibration data and embedded the models in a switching framework to detect transitions between states. We then created a density model (using a Gaussian mixture model) for the NLSSM innovations: this provides a model for normal operation. To validate our approach, we used data with added synthetic abnormalities which was detected as low-probability periods. The model of normality gave good indications of faults during the flight, in the form of low probabilities under the model, with high accuracy (>92 %). © 2013 IEEE.

Identifying PV module mismatch faults by a thermography-based temperature distribution analysis

Photovoltaic (PV) solar power generation is proven to be effective and sustainable but is currently hampered by relatively high costs and low conversion efficiency. This paper addresses both issues by presenting a low-cost and efficient temperature distribution analysis for identifying PV module mismatch faults by thermography. Mismatch faults reduce the power output and cause potential damage to PV cells. This paper first defines three fault categories in terms of fault levels, which lead to different terminal characteristics of the PV modules. The investigation of three faults is also conducted analytically and experimentally, and maintenance suggestions are also provided for different fault types. The proposed methodology is developed to combine the electrical and thermal characteristics of PV cells subjected to different fault mechanisms through simulation and experimental tests. Furthermore, the fault diagnosis method can be incorporated into the maximum power point tracking schemes to shift the operating point of the PV string. The developed technology has improved over the existing ones in locating the faulty cell by a thermal camera, providing a remedial measure, and maximizing the power output under faulty conditions.

Simple approximate MAP inference for Dirichlet processes mixtures

The Dirichlet process mixture model (DPMM) is a ubiquitous, flexible Bayesian nonparametric statistical model. However, full probabilistic inference in this model is analytically intractable, so that computationally intensive techniques such as Gibbs sampling are required. As a result, DPMM-based methods, which have considerable potential, are restricted to applications in which computational resources and time for inference is plentiful. For example, they would not be practical for digital signal processing on embedded hardware, where computational resources are at a serious premium. Here, we develop a simplified yet statistically rigorous approximate maximum a-posteriori (MAP) inference algorithm for DPMMs. This algorithm is as simple as DP-means clustering, solves the MAP problem as well as Gibbs sampling, while requiring only a fraction of the computational effort. (For freely available code that implements the MAP-DP algorithm for Gaussian mixtures see http://www.maxlittle.net/.) Unlike related small variance asymptotics (SVA), our method is non-degenerate and so inherits the “rich get richer” property of the Dirichlet process. It also retains a non-degenerate closed-form likelihood which enables out-of-sample calculations and the use of standard tools such as cross-validation. We illustrate the benefits of our algorithm on a range of examples and contrast it to variational, SVA and sampling approaches from both a computational complexity perspective as well as in terms of clustering performance. We demonstrate the wide applicabiity of our approach by presenting an approximate MAP inference method for the infinite hidden Markov model whose performance contrasts favorably with a recently proposed hybrid SVA approach. Similarly, we show how our algorithm can applied to a semiparametric mixed-effects regression model where the random effects distribution is modelled using an infinite mixture model, as used in longitudinal progression modelling in population health science. Finally, we propose directions for future research on approximate MAP inference in Bayesian nonparametrics.

Improved inference in the evaluation of mutual fund performance using bootstrap methods

Two new methodologies are introduced to improve inference in the evaluation of mutual fund performance against benchmarks. First, the benchmark models are estimated using panel methods with both fund and time effects. Second, the non-normality of individual mutual fund returns is accounted for by using panel bootstrap methods. We also augment the standard benchmark factors with fund-specific characteristics, such as fund size. Using a dataset of UK equity mutual fund returns, we find that fund size has a negative effect on the average fund manager’s benchmark-adjusted performance. Further, when we allow for time effects and the non-normality of fund returns, we find that there is no evidence that even the best performing fund managers can significantly out-perform the augmented benchmarks after fund management charges are taken into account.