Viability Kernels of Higher Order

AMS subject classification: Primary 34A60, Secondary 49K24.

Viability, Optimality and Stability of Dynamical Systems and Estimation of Convergence of Numerical Schemes

AMS subject classification: 49N35,49N55,65Lxx.

Introduction à la théorie de la viabilité

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

Survivability of Deterministic Dynamical Systems

Acknowledgements We acknowledge gratefully the support of BMBF, CoNDyNet, FK. 03SF0472A, of the EIT Climate-KIC project SWIPO and Nora Molkenthin for illustrating our illustration of the concept of survivability using penguins. We thank Martin Rohden for providing us with the UK high-voltage transmission grid topology and Yang Tang for very useful discussions. The publication of this article was funded by the Open Access Fund of the Leibniz Association.

An evaluation of viability of public private partnerships in social infrastructure procurement projects in Queensland, Australia

Public private partnerships (PPP) have been widely used as a method for public infrastructure project delivery not only locally and internationally, however the adoption of PPPs in social infrastructure procurement has still been very limited. The objective of this paper is to investigate the potential of implementation of current PPP framework in social affordable housing projects in South East Queensland. Data were collected from 22 interviewees with rich experiences in the industry. The findings of this study show that affordable housing investment have been considered by the industry practitioners as a risky business in comparison to other private rental housing investment. The main determents of the adoption of PPPs in social infrastructure project are the tenant-related factors, such as the inability of paying rent and the inability of caring the property. The study also suggests the importance of seeking strategic partnership with community-based organisation that has experiences in managing similar tenants’ profiles. Current PPP guideline is also viewed as inappropriate for the affordable housing projects, but the principle of VFM framework and risk allocation in PPPs still be applied to the affordable housing projects. This study helps to understand the viability of PPP in social housing procurement projects, and point out the importance of developing guideline for multi-stakeholder partnership and the expansion of the current VFM and PPPs guidelines.

If it's encrypted it's secure! The viability of US state-based encryption exemptions

US state-based data breach notification laws have unveiled serious corporate and government failures regarding the security of personal information. These laws require organisations to notify persons who may be affected by an unauthorized acquisition of their personal information. Safe harbours to notification exist if personal information is encrypted. Three types of safe harbour have been identified in the literature: exemptions, rebuttable presumptions and factors. The underlying assumption of exemptions is that encrypted personal information is secure and therefore unauthorized access does not pose a risk. However, the viability of this assumption is questionable when examined against data breaches involving encrypted information and the demanding practical requirements of effective encryption management. Recent recommendations by the Australian Law Reform Commission (ALRC) would amend the Privacy Act 1988 (Cth) to implement a data breach scheme that includes a different type of safe harbour, factor based analysis. The authors examine the potential capability of the ALRC’s proposed encryption safe harbour in relation to the US experience at the state legislature level.

Color and texture feature fusion using kernel PCA with application to object-based vegetation species classification

A good object representation or object descriptor is one of the key issues in object based image analysis. To effectively fuse color and texture as a unified descriptor at object level, this paper presents a novel method for feature fusion. Color histogram and the uniform local binary patterns are extracted from arbitrary-shaped image-objects, and kernel principal component analysis (kernel PCA) is employed to find nonlinear relationships of the extracted color and texture features. The maximum likelihood approach is used to estimate the intrinsic dimensionality, which is then used as a criterion for automatic selection of optimal feature set from the fused feature. The proposed method is evaluated using SVM as the benchmark classifier and is applied to object-based vegetation species classification using high spatial resolution aerial imagery. Experimental results demonstrate that great improvement can be achieved by using proposed feature fusion method.

Differences between in vitro viability and differentiation and in vivo bone-forming efficacy of human mesenchymal stem cells cultured on PCL-TCP scaffolds

Human mesenchymal stem cells (hMSCs) possess great therapeutic potential for the treatment of bone disease and fracture non-union. Too often however, in vitro evidence alone of the interaction between hMSCs and the biomaterial of choice is used as justification for continued development of the material into the clinic. Clearly for hMSC-based regenerative medicine to be successful for the treatment of orthopaedic trauma, it is crucial to transplant hMSCs with a suitable carrier that facilitates their survival, optimal proliferation and osteogenic differentiation in vitro and in vivo. This motivated us to evaluate the use of polycaprolactone-20% tricalcium phosphate (PCL-TCP) scaffolds produced by fused deposition modeling for the delivery of hMSCs. When hMSCs were cultured on the PCL-TCP scaffolds and imaged by a combination of phase contrast, scanning electron and confocal laser microscopy, we observed five distinct stages of colonization over a 21-day period that were characterized by cell attachment, spreading, cellular bridging, the formation of a dense cellular mass and the accumulation of a mineralized extracellular matrix when induced with osteogenic stimulants. Having established that PCL-TCP scaffolds are able to support hMSC proliferation and osteogenic differentiation, we next tested the in vivo efficacy of hMSC-loaded PCL-TCP scaffolds in nude rat critical-sized femoral defects. We found that fluorescently labeled hMSCs survived in the defect site for up to 3 weeks post-transplantation. However, only 50% of the femoral defects treated with hMSCs responded favorably as determined by new bone volume. As such, we show that verification of hMSC viability and differentiation in vitro is not sufficient to predict the efficacy of transplanted stem cells to consistently promote bone formation in orthotopic defects in vivo.

Learning the kernel matrix with semidefinite programming

Kernel-based learning algorithms work by embedding the data into a Euclidean space, and then searching for linear relations among the embedded data points. The embedding is performed implicitly, by specifying the inner products between each pair of points in the embedding space. This information is contained in the so-called kernel matrix, a symmetric and positive semidefinite matrix that encodes the relative positions of all points. Specifying this matrix amounts to specifying the geometry of the embedding space and inducing a notion of similarity in the input space - classical model selection problems in machine learning. In this paper we show how the kernel matrix can be learned from data via semidefinite programming (SDP) techniques. When applied to a kernel matrix associated with both training and test data this gives a powerful transductive algorithm -using the labeled part of the data one can learn an embedding also for the unlabeled part. The similarity between test points is inferred from training points and their labels. Importantly, these learning problems are convex, so we obtain a method for learning both the model class and the function without local minima. Furthermore, this approach leads directly to a convex method for learning the 2-norm soft margin parameter in support vector machines, solving an important open problem.

A unifying view of multiple kernel learning

Recent research on multiple kernel learning has lead to a number of approaches for combining kernels in regularized risk minimization. The proposed approaches include different formulations of objectives and varying regularization strategies. In this paper we present a unifying optimization criterion for multiple kernel learning and show how existing formulations are subsumed as special cases. We also derive the criterion’s dual representation, which is suitable for general smooth optimization algorithms. Finally, we evaluate multiple kernel learning in this framework analytically using a Rademacher complexity bound on the generalization error and empirically in a set of experiments.

A unifying view of multiple kernel learning

Recent research on multiple kernel learning has lead to a number of approaches for combining kernels in regularized risk minimization. The proposed approaches include different formulations of objectives and varying regularization strategies. In this paper we present a unifying general optimization criterion for multiple kernel learning and show how existing formulations are subsumed as special cases. We also derive the criterion's dual representation, which is suitable for general smooth optimization algorithms. Finally, we evaluate multiple kernel learning in this framework analytically using a Rademacher complexity bound on the generalization error and empirically in a set of experiments.

Learning the Kernel Matrix with Semi-Definite Programming

Kernel-based learning algorithms work by embedding the data into a Euclidean space, and then searching for linear relations among the embedded data points. The embedding is performed implicitly, by specifying the inner products between each pair of points in the embedding space. This information is contained in the so-called kernel matrix, a symmetric and positive definite matrix that encodes the relative positions of all points. Specifying this matrix amounts to specifying the geometry of the embedding space and inducing a notion of similarity in the input space -- classical model selection problems in machine learning. In this paper we show how the kernel matrix can be learned from data via semi-definite programming (SDP) techniques. When applied to a kernel matrix associated with both training and test data this gives a powerful transductive algorithm -- using the labelled part of the data one can learn an embedding also for the unlabelled part. The similarity between test points is inferred from training points and their labels. Importantly, these learning problems are convex, so we obtain a method for learning both the model class and the function without local minima. Furthermore, this approach leads directly to a convex method to learn the 2-norm soft margin parameter in support vector machines, solving another important open problem. Finally, the novel approach presented in the paper is supported by positive empirical results.