Backward elimination model construction for regression and classification using leave-one-out criteria

A fundamental principle in practical nonlinear data modeling is the parsimonious principle of constructing the minimal model that explains the training data well. Leave-one-out (LOO) cross validation is often used to estimate generalization errors by choosing amongst different network architectures (M. Stone, "Cross validatory choice and assessment of statistical predictions", J. R. Stast. Soc., Ser. B, 36, pp. 117-147, 1974). Based upon the minimization of LOO criteria of either the mean squares of LOO errors or the LOO misclassification rate respectively, we present two backward elimination algorithms as model post-processing procedures for regression and classification problems. The proposed backward elimination procedures exploit an orthogonalization procedure to enable the orthogonality between the subspace as spanned by the pruned model and the deleted regressor. Subsequently, it is shown that the LOO criteria used in both algorithms can be calculated via some analytic recursive formula, as derived in this contribution, without actually splitting the estimation data set so as to reduce computational expense. Compared to most other model construction methods, the proposed algorithms are advantageous in several aspects; (i) There are no tuning parameters to be optimized through an extra validation data set; (ii) The procedure is fully automatic without an additional stopping criteria; and (iii) The model structure selection is directly based on model generalization performance. The illustrative examples on regression and classification are used to demonstrate that the proposed algorithms are viable post-processing methods to prune a model to gain extra sparsity and improved generalization.

A helmet mounted display system with active gaze control for visual telepresence

The objective of a Visual Telepresence System is to provide the operator with a high fidelity image from a remote stereo camera pair linked to a pan/tilt device such that the operator may reorient the camera position by use of head movement. Systems such as these which utilise virtual reality style helmet mounted displays have a number of limitations. The geometry of the camera positions and of the displays is generally fixed and is most suitable only for viewing elements of a scene at a particular distance. To address such limitations, a prototype system has been developed where the geometry of the displays and cameras is dynamically controlled by the eye movement of the operator. This paper explores why it is necessary to actively adjust the display system as well as the cameras and justifies the use of mechanical adjustment of the displays as an alternative to adjustment by electronic or image processing methods. The electronic and mechanical design is described including optical arrangements and control algorithms. The performance and accuracy of the system is assessed with respect to eye movement.

Dynamic vergence telepresence system incorporating active gaze control

Visual Telepresence system which utilize virtual reality style helmet mounted displays have a number of limitations. The geometry of the camera positions and of the display is fixed and is most suitable only for viewing elements of a scene at a particular distance. In such a system, the operator's ability to gaze around without use of head movement is severely limited. A trade off must be made between a poor viewing resolution or a narrow width of viewing field. To address these limitations a prototype system where the geometry of the displays and cameras is dynamically controlled by the eye movement of the operator has been developed. This paper explores the reasons why is necessary to actively adjust both the display system and the cameras and furthermore justifies the use of mechanical adjustment of the displays as an alternative to adjustment by electronic or image processing methods. The electronic and mechanical design is described including optical arrangements and control algorithms, An assessment of the performance of the system against a fixed camera/display system when operators are assigned basic tasks involving depth and distance/size perception. The sensitivity to variations in transient performance of the display and camera vergence is also assessed.

Deriving global flood hazard maps of fluvial floods through a physical model cascade

Global flood hazard maps can be used in the assessment of flood risk in a number of different applications, including (re)insurance and large scale flood preparedness. Such global hazard maps can be generated using large scale physically based models of rainfall-runoff and river routing, when used in conjunction with a number of post-processing methods. In this study, the European Centre for Medium Range Weather Forecasts (ECMWF) land surface model is coupled to ERA-Interim reanalysis meteorological forcing data, and resultant runoff is passed to a river routing algorithm which simulates floodplains and flood flow across the global land area. The global hazard map is based on a 30 yr (1979–2010) simulation period. A Gumbel distribution is fitted to the annual maxima flows to derive a number of flood return periods. The return periods are calculated initially for a 25×25 km grid, which is then reprojected onto a 1×1 km grid to derive maps of higher resolution and estimate flooded fractional area for the individual 25×25 km cells. Several global and regional maps of flood return periods ranging from 2 to 500 yr are presented. The results compare reasonably to a benchmark data set of global flood hazard. The developed methodology can be applied to other datasets on a global or regional scale.

On the reprocessing and reanalysis of observations for climate

The long observational record is critical to our understanding of the Earth’s climate, but most observing systems were not developed with a climate objective in mind. As a result, tremendous efforts have gone into assessing and reprocessing the data records to improve their usefulness in climate studies. The purpose of this paper is to both review recent progress in reprocessing and reanalyzing observations, and summarize the challenges that must be overcome in order to improve our understanding of climate and variability. Reprocessing improves data quality through more scrutiny and improved retrieval techniques for individual observing systems, while reanalysis merges many disparate observations with models through data assimilation, yet both aim to provide a climatology of Earth processes. Many challenges remain, such as tracking the improvement of processing algorithms and limited spatial coverage. Reanalyses have fostered significant research, yet reliable global trends in many physical fields are not yet attainable, despite significant advances in data assimilation and numerical modeling. Oceanic reanalyses have made significant advances in recent years, but will only be discussed here in terms of progress toward integrated Earth system analyses. Climate data sets are generally adequate for process studies and large-scale climate variability. Communication of the strengths, limitations and uncertainties of reprocessed observations and reanalysis data, not only among the community of developers, but also with the extended research community, including the new generations of researchers and the decision makers is crucial for further advancement of the observational data records. It must be emphasized that careful investigation of the data and processing methods are required to use the observations appropriately.

Computer intensive methods in control and signal processing: the curse of dimensionality

This volume is based upon the 2nd IEEE European Workshop on Computer-Intensive Methods in Control and Signal Processing, held in Prague, August 1996.

Approximate Gauss-Newton methods for nonlinear least squares problems

The Gauss–Newton algorithm is an iterative method regularly used for solving nonlinear least squares problems. It is particularly well suited to the treatment of very large scale variational data assimilation problems that arise in atmosphere and ocean forecasting. The procedure consists of a sequence of linear least squares approximations to the nonlinear problem, each of which is solved by an “inner” direct or iterative process. In comparison with Newton’s method and its variants, the algorithm is attractive because it does not require the evaluation of second-order derivatives in the Hessian of the objective function. In practice the exact Gauss–Newton method is too expensive to apply operationally in meteorological forecasting, and various approximations are made in order to reduce computational costs and to solve the problems in real time. Here we investigate the effects on the convergence of the Gauss–Newton method of two types of approximation used commonly in data assimilation. First, we examine “truncated” Gauss–Newton methods where the inner linear least squares problem is not solved exactly, and second, we examine “perturbed” Gauss–Newton methods where the true linearized inner problem is approximated by a simplified, or perturbed, linear least squares problem. We give conditions ensuring that the truncated and perturbed Gauss–Newton methods converge and also derive rates of convergence for the iterations. The results are illustrated by a simple numerical example. A practical application to the problem of data assimilation in a typical meteorological system is presented.

A geometric algorithm for an artificial life form based on anachronistic methods

Modern methods of spawning new technological motifs are not appropriate when it is desired to realize artificial life as an actual real world entity unto itself (Pattee 1995; Brooks 2006; Chalmers 1995). Many fundamental aspects of such a machine are absent in common methods, which generally lack methodologies of construction. In this paper we mix classical and modern studies in order to attempt to realize an artificial life form from first principles. A model of an algorithm is introduced, its methodology of construction is presented, and the fundamental source from which it sprang is discussed.

On-line processing of wh-questions in children with G-SLI and typically developing children

Background: The computational grammatical complexity ( CGC) hypothesis claims that children with G(rammatical)-specific language impairment ( SLI) have a domain-specific deficit in the computational system affecting syntactic dependencies involving 'movement'. One type of such syntactic dependencies is filler-gap dependencies. In contrast, the Generalized Slowing Hypothesis claims that SLI children have a domain-general deficit affecting processing speed and capacity. Aims: To test contrasting accounts of SLI we investigate processing of syntactic (filler-gap) dependencies in wh-questions. Methods & Procedures: Fourteen 10; 2 - 17; 2 G-SLI children, 14 age- matched and 17 vocabulary-matched controls were studied using the cross- modal picturepriming paradigm. Outcomes & Results: G-SLI children's processing speed was significantly slower than the age controls, but not younger vocabulary controls. The G- SLI children and vocabulary controls did not differ on memory span. However, the typically developing and G-SLI children showed a qualitatively different processing pattern. The age and vocabulary controls showed priming at the gap, indicating that they process wh-questions through syntactic filler-gap dependencies. In contrast, G-SLI children showed priming only at the verb. Conclusions: The findings indicate that G-SLI children fail to establish reliably a syntactic filler- gap dependency and instead interpret wh-questions via lexical thematic information. These data challenge the Generalized Slowing Hypothesis account, but support the CGC hypothesis, according to which G-SLI children have a particular deficit in the computational system affecting syntactic dependencies involving 'movement'. As effective remediation often depends on aetiological insight, the discovery of the nature of the syntactic deficit, along side a possible compensatory use of semantics to facilitate sentence processing, can be used to direct therapy. However, the therapeutic strategy to be used, and whether such similar strengths and weaknesses within the language system are found in other SLI subgroups are empirical issues that warrant further research.