Structured output SVM for remote sensing image classification

In the recent years, kernel methods have revealed very powerful tools in many application domains in general and in remote sensing image classification in particular. The special characteristics of remote sensing images (high dimension, few labeled samples and different noise sources) are efficiently dealt with kernel machines. In this paper, we propose the use of structured output learning to improve remote sensing image classification based on kernels. Structured output learning is concerned with the design of machine learning algorithms that not only implement input-output mapping, but also take into account the relations between output labels, thus generalizing unstructured kernel methods. We analyze the framework and introduce it to the remote sensing community. Output similarity is here encoded into SVM classifiers by modifying the model loss function and the kernel function either independently or jointly. Experiments on a very high resolution (VHR) image classification problem shows promising results and opens a wide field of research with structured output kernel methods.

Haar wavelets-based approach for quantifying credit portfolio losses

This paper proposes a new methodology to compute Value at Risk (VaR) for quantifying losses in credit portfolios. We approximate the cumulative distribution of the loss function by a finite combination of Haar wavelet basis functions and calculate the coefficients of the approximation by inverting its Laplace transform. The Wavelet Approximation (WA) method is specially suitable for non-smooth distributions, often arising in small or concentrated portfolios, when the hypothesis of the Basel II formulas are violated. To test the methodology we consider the Vasicek one-factor portfolio credit loss model as our model framework. WA is an accurate, robust and fast method, allowing to estimate VaR much more quickly than with a Monte Carlo (MC) method at the same level of accuracy and reliability.

A Log-linear model of grain size influence on the geochemistry of sediments

Sediment composition is mainly controlled by the nature of the source rock(s), and chemical (weathering) and physical processes (mechanical crushing, abrasion, hydrodynamic sorting) during alteration and transport. Although the factors controlling these processes are conceptually well understood, detailed quantification of compositional changes induced by a single process are rare, as are examples where the effects of several processes can be distinguished. The present study was designed to characterize the role of mechanical crushing and sorting in the absence of chemical weathering. Twenty sediment samples were taken from Alpine glaciers that erode almost pure granitoid lithologies. For each sample, 11 grain-size fractions from granules to clay (ø grades &-1 to &9) were separated, and each fraction was analysed for its chemical composition.The presence of clear steps in the box-plots of all parts (in adequate ilr and clr scales) against ø is assumed to be explained by typical crystal size ranges for the relevant mineral phases. These scatter plots and the biplot suggest a splitting of the full grain size range into three groups: coarser than ø=4 (comparatively rich in SiO2, Na2O, K2O, Al2O3, and dominated by “felsic” minerals like quartz and feldspar), finer than ø=8 (comparatively rich in TiO2, MnO, MgO, Fe2O3, mostly related to “mafic” sheet silicates like biotite and chlorite), and intermediate grains sizes (4≤ø &8; comparatively rich in P2O5 and CaO, related to apatite, some feldspar).To further test the absence of chemical weathering, the observed compositions were regressed against three explanatory variables: a trend on grain size in ø scale, a step function for ø≥4, and another for ø≥8. The original hypothesis was that the trend could be identified with weathering effects, whereas each step function would highlight those minerals with biggest characteristic size at its lower end. Results suggest that this assumption is reasonable for the step function, but that besides weathering some other factors (different mechanical behavior of minerals) have also an important contribution to the trend.Key words: sediment, geochemistry, grain size, regression, step function

Which inflation to target? A small open economy with sticky wages indexed to past inflation

In a closed economy context there is common agreement on price inflation stabilization being one of the objects of monetary policy. Moving to an open economy context gives rise to the coexistence of two measures of inflation: domestic inflation (DI) and consumer price inflation (CPI). Which one of the two measures should be the target variable? This is the question addressed in this paper. In particular, I use a small open economy model to show that once sticky wages indexed to past CPI inflation are introduced, a complete inward looking monetary policy is no more optimal. I first, derive a loss function from a secondorder approximation of the utility function and then, I compute the fully optimalmonetary policy under commitment. Then, I use the optimal monetary policy as a benchmark to compare the performance of different monetary policy rules. The main result is that once a positive degree of indexation is introduced in the model the rule performing better (among the Taylor type rules considered) is the one targeting wage inflation and CPI inflation. Moreover this rule delivers results very close to the one obtained under the fully optimal monetary policy with commitment.

Optical response of Cu3Ge thin films

We report an investigation on the optical properties of Cu3Ge thin films displaying very high conductivity, with thickness ranging from 200 to 2000 Å, deposited on Ge substrates. Reflectance, transmittance, and ellipsometric spectroscopy measurements were performed at room temperature in the 0.01-6.0, 0.01-0.6, and 1.4-5.0 eV energy range, respectively. The complex dielectric function, the optical conductivity, the energy-loss function, and the effective charge density were obtained over the whole spectral range. The low-energy free-carrier response was well fitted by using the classical Drude-Lorentz dielectric function. A simple two-band model allowed the resulting optical parameters to be interpreted coherently with those previously obtained from transport measurements, hence yielding the densities and the effective masses of electrons and holes.

The database search problem: a question of rational decision making

This paper applies probability and decision theory in the graphical interface of an influence diagram to study the formal requirements of rationality which justify the individualization of a person found through a database search. The decision-theoretic part of the analysis studies the parameters that a rational decision maker would use to individualize the selected person. The modeling part (in the form of an influence diagram) clarifies the relationships between this decision and the ingredients that make up the database search problem, i.e., the results of the database search and the different pairs of propositions describing whether an individual is at the source of the crime stain. These analyses evaluate the desirability associated with the decision of 'individualizing' (and 'not individualizing'). They point out that this decision is a function of (i) the probability that the individual in question is, in fact, at the source of the crime stain (i.e., the state of nature), and (ii) the decision maker's preferences among the possible consequences of the decision (i.e., the decision maker's loss function). We discuss the relevance and argumentative implications of these insights with respect to recent comments in specialized literature, which suggest points of view that are opposed to the results of our study.

On the relationship between connections and the asymptotic properties of predictive distributions

In a recent paper, Komaki studied the second-order asymptotic properties of predictive distributions, using the Kullback-Leibler divergence as a loss function. He showed that estimative distributions with asymptotically efficient estimators can be improved by predictive distributions that do not belong to the model. The model is assumed to be a multidimensional curved exponential family. In this paper we generalize the result assuming as a loss function any f divergence. A relationship arises between alpha connections and optimal predictive distributions. In particular, using an alpha divergence to measure the goodness of a predictive distribution, the optimal shift of the estimate distribution is related to alpha-covariant derivatives. The expression that we obtain for the asymptotic risk is also useful to study the higher-order asymptotic properties of an estimator, in the mentioned class of loss functions.

Applying life cycle management of colombian cocoa production

The present research aims to evaluate the usefulness of the application of Life Cycle Management in the agricultural sector focusing on the environmental and socio-economic aspects of decision making in the Colombian cocoa production. Such appraisal is based on the application of two methodological tools: Life Cycle Assessment, which considers environmental impacts throughout the life cycle of the cocoa production system, and Taguchi Loss Function, which measures the economic impact of a process' deviation from production targets. Results show that appropriate improvements in farming practices and supply consumption can enhance decision-making in the agricultural cocoa sector towards sustainability. In terms of agri-business purposes, such qualitative shift allows not only meeting consumer demands for environmentally friendly products, but also increasing the productivity and competitiveness of cocoa production, all of which has helped Life Cycle Management gain global acceptance. Since farmers have an important role in improving social and economic indicators at the national level, more attention should be paid to the upgrading of their cropping practices. Finally, one fundamental aspect of national cocoa production is the institutional and governmental support available for farmers in face of socio-economic or technological needs.

Inflation Targeting Under Asymmetric Preferences

This paper develops and estimates a game-theoretical model of inflation targeting where the central banker's preferences are asymmetric around the targeted rate. In particular, positive deviations from the target can be weighted more, or less, severely than negative ones in the central banker's loss function. It is shown that some of the previous results derived under the assumption of symmetry are not robust to the generalization of preferences. Estimates of the central banker's preference parameters for Canada, Sweden, and the United Kingdom are statistically different from the ones implied by the commonly used quadratic loss function. Econometric results are robust to different forecasting models for the rate of unemployment but not to the use of measures of inflation broader than the one targeted.

A Prudent Central Banker

This paper studies monetary policy in an economy where the central banker's preferences are asymmetric around optimal inflation. In particular, positive deviations from the optimum can be weighted more, or less, severely than negative deviations in the policy maker's loss function. It is shown that under asymmetric preferences, uncertainty can induce a prudent behavior on the part of the central banker. Since the prudence motive can be large enough to override the inflation bias, optimal monetary policy could be implemented even in the absence of rules, reputation, or contractual mechanisms. For certain parameter values, a deflationary bias can arise in equilibrium.

Uncovering Financial Markets Beliefs About Inflation Targets

This paper exploits the term structure of interest rates to develop testable economic restrictions on the joint process of long-term interest rates and inflation when the latter is subject to a targeting policy by the Central Bank. Two competing models that econometrically describe agents’ inferences about inflation targets are developed and shown to generate distinct predictions on the behavior of interest rates. In an empirical application to the Canadian inflation target zone, results indicate that agents perceive the band to be substantially narrower than officially announced and asymmetric around the stated mid-point. The latter result (i) suggests that the monetary authority attaches different weights to positive and negative deviations from the central target, and (ii) challenges on empirical grounds the assumption, frequently made in the literature, that the policy maker’s loss function is symmetric (usually a quadratic function) around a desired inflation value.

The Inflation Bias When the Central Bank Targets, the Natural Rate of Unemployment

This paper studies the proposition that an inflation bias can arise in a setup where a central banker with asymmetric preferences targets the natural unemployment rate. Preferences are asymmetric in the sense that positive unemployment deviations from the natural rate are weighted more (or less) severely than negative deviations in the central banker's loss function. The bias is proportional to the conditional variance of unemployment. The time-series predictions of the model are evaluated using data from G7 countries. Econometric estimates support the prediction that the conditional variance of unemployment and the rate of inflation are positively related.

Fonctions de perte en actuariat

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

On the Noise Model of Support Vector Machine Regression

Support Vector Machines Regression (SVMR) is a regression technique which has been recently introduced by V. Vapnik and his collaborators (Vapnik, 1995; Vapnik, Golowich and Smola, 1996). In SVMR the goodness of fit is measured not by the usual quadratic loss function (the mean square error), but by a different loss function called Vapnik"s $epsilon$- insensitive loss function, which is similar to the "robust" loss functions introduced by Huber (Huber, 1981). The quadratic loss function is well justified under the assumption of Gaussian additive noise. However, the noise model underlying the choice of Vapnik's loss function is less clear. In this paper the use of Vapnik's loss function is shown to be equivalent to a model of additive and Gaussian noise, where the variance and mean of the Gaussian are random variables. The probability distributions for the variance and mean will be stated explicitly. While this work is presented in the framework of SVMR, it can be extended to justify non-quadratic loss functions in any Maximum Likelihood or Maximum A Posteriori approach. It applies not only to Vapnik's loss function, but to a much broader class of loss functions.

On the V(subscript gamma) Dimension for Regression in Reproducing Kernel Hilbert Spaces

This paper presents a computation of the $V_gamma$ dimension for regression in bounded subspaces of Reproducing Kernel Hilbert Spaces (RKHS) for the Support Vector Machine (SVM) regression $epsilon$-insensitive loss function, and general $L_p$ loss functions. Finiteness of the RV_gamma$ dimension is shown, which also proves uniform convergence in probability for regression machines in RKHS subspaces that use the $L_epsilon$ or general $L_p$ loss functions. This paper presenta a novel proof of this result also for the case that a bias is added to the functions in the RKHS.