Adaptive model selection using empirical complexities

Given $n$ independent replicates of a jointly distributed pair $(X,Y)\in {\cal R}^d \times {\cal R}$, we wish to select from a fixed sequence of model classes ${\cal F}_1, {\cal F}_2, \ldots$ a deterministic prediction rule $f: {\cal R}^d \to {\cal R}$ whose risk is small. We investigate the possibility of empirically assessingthe {\em complexity} of each model class, that is, the actual difficulty of the estimation problem within each class. The estimated complexities are in turn used to define an adaptive model selection procedure, which is based on complexity penalized empirical risk.The available data are divided into two parts. The first is used to form an empirical cover of each model class, and the second is used to select a candidate rule from each cover based on empirical risk. The covering radii are determined empirically to optimize a tight upper bound on the estimation error. An estimate is chosen from the list of candidates in order to minimize the sum of class complexity and empirical risk. A distinguishing feature of the approach is that the complexity of each model class is assessed empirically, based on the size of its empirical cover.Finite sample performance bounds are established for the estimates, and these bounds are applied to several non-parametric estimation problems. The estimates are shown to achieve a favorable tradeoff between approximation and estimation error, and to perform as well as if the distribution-dependent complexities of the model classes were known beforehand. In addition, it is shown that the estimate can be consistent,and even possess near optimal rates of convergence, when each model class has an infinite VC or pseudo dimension.For regression estimation with squared loss we modify our estimate to achieve a faster rate of convergence.

Revealed preference and indifferent selection

It is shown that preferences can be constructed from observed choice behavior in a way that is robust to indifferent selection (i.e., the agent is indifferent between two alternatives but, nevertheless, is only observed selecting one of them). More precisely, a suggestion by Savage (1954) to reveal indifferent selection by considering small monetary perturbations of alternatives is formalized and generalized to a purely topological framework: references over an arbitrary topological space can be uniquely derived from observed behavior under the assumptions that they are continuous and nonsatiated and that a strictly preferred alternative is always chosen, and indifferent selection is then characterized by discontinuity in choice behavior. Two particular cases are then analyzed: monotonic preferences over a partially ordered set, and preferences representable by a continuous pseudo-utility function.

Local selection rules that can determine specific pathways of DNA unknotting by type II DNA topoisomerases.

We performed numerical simulations of DNA chains to understand how local geometry of juxtaposed segments in knotted DNA molecules can guide type II DNA topoisomerases to perform very efficient relaxation of DNA knots. We investigated how the various parameters defining the geometry of inter-segmental juxtapositions at sites of inter-segmental passage reactions mediated by type II DNA topoisomerases can affect the topological consequences of these reactions. We confirmed the hypothesis that by recognizing specific geometry of juxtaposed DNA segments in knotted DNA molecules, type II DNA topoisomerases can maintain the steady-state knotting level below the topological equilibrium. In addition, we revealed that a preference for a particular geometry of juxtaposed segments as sites of strand-passage reaction enables type II DNA topoisomerases to select the most efficient pathway of relaxation of complex DNA knots. The analysis of the best selection criteria for efficient relaxation of complex knots revealed that local structures in random configurations of a given knot type statistically behave as analogous local structures in ideal geometric configurations of the corresponding knot type.

A model of collateral, investment and adverse selection

This paper characterizes the relationship between entrepreneurial wealth and aggregate investmentunder adverse selection. Its main finding is that such a relationship need not bemonotonic. In particular, three results emerge from the analysis: (i) pooling equilibria, in whichinvestment is independent of entrepreneurial wealth, are more likely to arise when entrepreneurialwealth is relatively low; (ii) separating equilibria, in which investment is increasing inentrepreneurial wealth, are most likely to arise when entrepreneurial wealth is relatively highand; (iii) for a given interest rate, an increase in entrepreneurial wealth may generate a discontinuousfall in investment.

Model selection and error estimation

We study model selection strategies based on penalized empirical loss minimization. We point out a tight relationship between error estimation and data-based complexity penalization: any good error estimate may be converted into a data-based penalty function and the performance of the estimate is governed by the quality of the error estimate. We consider several penalty functions, involving error estimates on independent test data, empirical {\sc vc} dimension, empirical {\sc vc} entropy, andmargin-based quantities. We also consider the maximal difference between the error on the first half of the training data and the second half, and the expected maximal discrepancy, a closely related capacity estimate that can be calculated by Monte Carlo integration. Maximal discrepancy penalty functions are appealing for pattern classification problems, since their computation is equivalent to empirical risk minimization over the training data with some labels flipped.

Egg laying site selection by a host plant specialist leaf miner moth at two intra-plant levels in the northern Chilean Atacama Desert

Egg laying site selection by a host plant specialist leaf miner moth at two intra-plant levels in the northern Chilean Atacama Desert. The spatial distribution of the immature stages of the leaf miner Angelabella tecomae Vargas & Parra, 2005 was determined at two intra-plant levels (shoot and leaflet) on the shrub Tecoma fulva fulva (Cav.) D. Don (Bignoniaceae) in the Azapa valley, northern Chilean Atacama Desert. An aggregated spatial pattern was detected for all the immature stages along the shoot, with an age dependent relative position: eggs and first instar larvae were clumped at apex; second, third and fourth instar larvae were mostly found at intermediate positions; meanwhile the spinning larva and pupa were clumped at basis. This pattern suggests that the females select new, actively growing leaflets for egg laying. At the leaflet level, the immature stages were found more frequently at underside. Furthermore, survivorship was higher for larvae from underside mines. All these results highlight the importance of an accurate selection of egg laying site in the life history of this highly specialized leaf miner. By contrast, eventual wrong choices in the egg laying site selection may be associated with diminished larval survivorship. The importance of the continuous availability of new plant tissue in this highly human modified arid environment is discussed in relation with the observed patterns.

Agreed upon procedures of the Iowa Department of Human Services, IowaCare Project for the year ended June 30, 2008

Agreed upon procedures of the Iowa Department of Human Services, IowaCare Project for the year ended June 30, 2008

Revised NEO personality inventory normative data for catalan police officer selection: preliminary study

El Servei d'Avaluació, Seguiment i Selecció de l'ISPC han elaborat un estudi sobre el perfil de personalitat dels aspirants al Curs de Formació bàsica per policies, que es va presentar a l'International Society for the Study of Individual Differences Meeting celebrat al CosmoCaixa de Barcelona i que organitzen conjuntament l’Associació Iberoamericana per a la recerca de les diferències individuals i la Universitat de Barcelona. L’estudi, titulat Revised NEO Personality Inventory Normative Data for Catalan police officer selection: A preliminary study, té com a objectiu comparar els perfils de personalitat d’una mostra d’aspirants de l’ISPC amb els resultats d’una mostra d’aspirants a policia dels EUA, publicada en una revista científica de prestigi el mes de febrer passat. Els resultats mostren que els aspirants catalans destaquen per obtenir millors puntuacions en les dimensions de responsabilitat i amabilitat, cosa que indicaria que aquest tret es valora especialment durant el procés de selecció de la policia de Catalunya; en altres característiques de la personalitat les dues mostres obtenen resultats similars. Els trets característics del perfil del policia català seria el de persones estables emocionalment, poc impulsives, amb capacitat per gestionar l’estrés, orientades a les persones, agradables, sociables, responsables, disciplinades i cauteloses. Enllaç a: International Society for the Study of Individual Differences Meeting :http://www.issid.org/conferences/ISSID2013/ISSIDconference2013.html

Un projet multicentrique suisse pour améliorer la prévention et la récidive après un syndrome coronarien aigu [A Swiss multicentric project to improve the prevention of cardiovascular event recurrence after acute coronary syndromes].

Recurrence of cardiovascular events and mortality remain high after acute coronary syndromes. A Swiss multicentric study, "Inflammation and acute coronary syndromes (ACS)--Novel strategies for prevention and clinical managements", is currently underway with the support of the Swiss National Science Foundation. The study includes a clinical research subproject of which the aim is to assess the impact of the ELIPS program (multi-dimEnsionaL prevention Program after acute coronary Syndrome) on the recurrence of cardiovascular events after an ACS. The basic research sub-projects aim to investigate novel cardiovascular risk biomarkers and genetic determinants of recurrence and to study the role of stem cells after an ACS. Another sub-project will evaluate intracoronary imaging techniques and the efficacy of different types of stents.

Evaluation of an online, case-based interactive approach to teaching pathophysiology

Rapport de synthèse : L'article qui fait l'objet de ma thèse évalue une nouvelle approche pédagogique pour l'apprentissage de certains chapitres de physiopathologie. Le dispositif pédagogique se base sur l'alternance d'apprentissage ex-cathedra et de l'utilisation d'un site web comprenant des vignettes cliniques. Lors de la consultation de ces-dernières, l'étudiant est invité à demander des examens de laboratoire dont il pourrait justifier la pertinence selon le cas clinique étudié. La nouveauté du procédé réside dans le fait que, préalablement à son cours ex-cathedra, l'enseignant peut consulter les statistiques de demandes de laboratoire et ainsi orienter son cours selon les éléments mal compris par les étudiants. A la suite du cours ex-cathedra, les étudiants peuvent consulter sur internet la vignette clinique complète avec des explications. A l'issue de tout le cours, une évaluation auprès des étudiants a été conduite. Le procédé a été mis en place durant deux années consécutives et l'article en discute notamment les résultats. Nous avons pu conclure que cette méthode innovatrice d'enseignement amène les étudiants à mieux se préparer pour les cours ex-cathedra tout en permettant à l'enseignant d'identifier plus précisément quelles thématiques étaient difficiles pour les étudiants et donc d'ajuster au mieux son cours. Mon travail de thèse a consisté à créer ce dispositif d'apprentissage, à créer l'application web des vignettes cliniques et à l'implanter durant deux années consécutives. J'ai ensuite analysé les données des évaluations et écrit l'article que j'ai présenté à la revue 'Medical Teacher'. Après quelques corrections et précisions demandées par le comité de lecture, l'article a été accepté et publié. Ce travail a débouché sur une seconde version de l'application web qui est actuellement utilisée lors du module 3.1 de 3è année à l'Ecole de Médecine à Lausanne. Summary : Since the early days of sexual selection, our understanding of the selective forces acting on males and females during reproduction has increased remarkably. However, despite a long tradition of experimental and theoretical work in this field and relentless effort, numerous questions remain unanswered and many results are conflicting. Moreover, the interface between sexual selection and conservation biology has to date received little attention, despite existing evidence for its importance. In the present thesis, I first used an empirical approach to test various sexual selection hypotheses in a population of whitefish of central Switzerland. This precise population is characterized by a high prevalence of gonadal alterations in males. In particular, I challenged the hypothesis that whitefish males displaying peculiar gonadal features are of lower genetic quality than other seemingly normal males. Additionally, I also worked on identifying important determinant of sperm behavior. During a second theoretical part of my work, which is part of a larger project on the evolution of female mate preferences in harvested fish populations, I developed an individual-based simulation model to estimate how different mate discrimination costs affect the demographical behavior of fish populations and the evolutionary trajectories of female mate preferences. This latter work provided me with some insight on a recently published article addressing the importance of sexual selection for harvesting-induced evolution. I built upon this insight in a short perspective paper. In parallel, I let some methodological questions drive my thoughts, and wrote an essay about possible synergies between the biological, the philosophical and the statistical approach to biological questions.

Dissociation of thymic positive and negative selection in transgenic mice expressing major histocompatibility complex class I molecules exclusively on thymic cortical epithelial cells.

Thymic positive and negative selection of developing T lymphocytes confronts us with a paradox: How can a T-cell antigen receptor (TCR)-major histocompatibility complex (MHC)/peptide interaction in the former process lead to transduction of signals allowing for cell survival and in the latter induce programmed cell death or a hyporesponsive state known as anergy? One of the hypotheses put forward states that the outcome of a TCR-MHC/peptide interaction depends on the cell type presenting the selecting ligand to the developing thymocyte. Here we describe the development and lack of self-tolerance of CD8(+) T lymphocytes in transgenic mice expressing MHC class I molecules in the thymus exclusively on cortical epithelial cells. Despite the absence of MHC class I expression on professional antigen-presenting cells, normal numbers of CD8(+) cells were observed in the periphery. Upon specific activation, transgenic CD8(+) T cells efficiently lysed syngeneic MHC class I(+) targets in vitro and in vivo, indicating that thymic cortical epithelium (in contrast to medullary epithelium and antigen-presenting cells of hematopoietic origin) is incapable of tolerance induction. Thus, compartmentalization of the antigen-presenting cells involved in thymic positive selection and tolerance induction can (at least in part) explain the positive/negative selection paradox.

1st radiometric dating of a paleontologically dated Bathonian level from Georgia (USSR) - Use of the cathodoluminescence for selection of suitable plagioclases

Jurassic volcanic formations interlayered with (ammonite-bearing) sediments are common in the Caucasus area; this situation is of interest for the numerical calibration of the poorly documented Jurassic portion of the time scale. However, following petrographic study on thin sections no whole-rocks can be considered reliable geochronometers due to subsequent alteration; from about 20 samples, two were selected for plagioclase dating; one (V134) is probably early Kimmeridgian in age; the other (V136) is probably located in the Lower Bathonian stage according to diagnostic ammonites. Cathodoluminescence (CTL) study has shown that sample V136 was similar to usual volcanic feldspars (blue to green colour); however, the lack of CTL of the V134 plagioclase is a character common to diagenetic feldspars; consequently, in spite of a good optical preservation, this geo-chronometer cannot give an age representative of the time of emplacement of the lava flow. We have combined CTL observation with microprobe analysis in order to document the poorly known CTL behaviour of volcanic feldspars; cations Ti4+ and Fe2+ play a major role in the CTL colour of plagioclases and are able to document the growing history of these feldspars ; phenocrysts are initially rich in Fe2+ (core of the crystals, green in colour), then richer in Ti toward the exterior; microcrysts are even richer in Ti (blue to bright blue). We have also observed that natural CTL colour was modified resulting from acid ``cleaning'' of the separated feldspars : the initial blue or green colour tends to change to yellow or violet, respectively, after acid treatment probably due to oxydation of Fe2+ toward Fe3+. X-ray and microprobe analyses both indicated that plagioclases from sample V134 was near the sodic end member (albite) suggesting a diagenetic origin in this andesitic basalt; In contrast, sample V136 contains a calcic plagioclase of common composition for a doleritic basalt. The K-Ar conventional technique was applied as a preliminary tool for radiometric analysis. The Kimmeridgian Na-plagioclase sample gave a ``rejuvenated'' (85 Ma) apparent age which confirms a late genesis for the separated plagioclase phase; this interpretation is based on CTL observation, X-ray analysis, and microprobe analysis ; these techniques are able to distinguish samples which have been submitted to diagenetic alteration from those which have not. An age consistent with the stratigraphic location has been obtained from sample V136. This age of 161 +/- 3 (2-sigma) Ma, is the first one available from a sample palaeontologically located with reasonable precision within the mid Jurassic time.

Contribution of major cardiovascular risk factors to familial premature coronary artery disease : the GENECARD project

Résumé Objectifs: Cette étude relève la prévalence des principaux facteurs de risque cardiovasculaire dans les coronaropathies précoces (P-CAD) familiales, survenant chez au moins deux frères et/ou soeurs d'une même fratrie. Méthodes: Nous avons recruté 213 survivants atteints de P-CAD, issus de 103 fratries, diagnostiqués avant l'âge de 50 ans chez les hommes et 55 ans chez les femmes. La présence ou non d'hypertension, d'hypercholestérolémie, d'obésité et de tabagisme a été documentée au moment de l'événement chez 163 de ces patients (145 hommes et 18 femmes). Chaque patient a été comparé à deux individus de même âge et sexe, chez qui un diagnostic de P-CAD «sporadique» (non familiale) était posé, et à trois individus choisis au hasard parmi la population générale. Résultats: En comparaison de la population générale, les patients atteints de P-CAD sporadique avaient une prévalence supérieure pour l 'hypertension (29% vs. 14%, p<0.001), le cholestérol (54% vs. 33%, p<0.001), l'obésité (20% vs. 13%, p<0.001) et le tabagisme (76% vs. 39%, p<0.001). Ces facteurs de risque étaient de prévalences similaires, voire supérieures chez les patients atteints de P-CAD familiale (43% [p0.05 vs. P-CAD sporadiques], 58% [p=0.07], 21% et 72% respectivement). Seulement 7 (4%) des 163 patients atteints de P-CAD familiale et 22 (7%) des 326 patients atteints de P-CAD sporadique, ne présentaient aucun facteur de risque cardiovasculaire, comparés à 167 (34%) des 489 patients issus de la population générale. Conclusions: Les facteurs de risque cardiovasculaire classiques et réversibles ont une haute prévalence chez les patients atteints de P-CAD familiale. Ce fait rend improbable une contribution génétique prédominante, agissant en l'absence de facteurs de risque. Summary Objectives: This study was designed to assess the prevalence of major cardiovascular risk factors in familial premature coronary artery disease (P-CAD), affecting two or more siblings within one sibship. Background: Premature CAD has a genetic component. It remains to be established whether familial P-CAD is due to genes acting independently from major cardiovascular risk factors. Methods: We recruited 213 P-CAD survivors from 103 sibships diagnosed before age ?50 (men) or ?55 (women) years old. Hypertension, hypercholesterolemia, obesity, and smoking were documented at the time of the event in 163 patients (145 men and 18 women). Each patient was compared with two individuals of the same age and gender, diagnosed with sporadic (nonfamilial) P-CAD, and three individuals randomly sampled from the general population. Result: Compared with the general population, patients with sporadic P-CAD had a higher prevalence of hypertension (29% vs. 14%, p < 0.001), hypercholesterolemia (54% vs. 33%, p < 0.001), obesity (20% vs. 13%, p < 0.01), and smoking (76% vs. 39%, p < 0.001). These risk factors were equally or even more prevalent in patients with familial P-CAD (43% [p < 0.05 vs. sporadic P-CAD], 58% [p = 0.07], 21% and 72%, respectively). Overall, only 7 (4%) of 163 of patients with familial P-CAD and 22 (7%) of 326 of patients with sporadic P-CAD had none of these conditions, as compared with 167 (34%) of 489 patients in the general population. Conclusions: Classic, remediable risk factors are highly prevalent in patients with familial P-CAD. Accordingly, a major contribution of genes acting in the absence of these risk factors is unlikely.