Self-healing umbrella sampling: convergence and efficiency

The Self-Healing Umbrella Sampling (SHUS) algorithm is an adaptive biasing algorithm which has been proposed in Marsili et al. (J Phys Chem B 110(29):14011–14013, 2006) in order to efficiently sample a multimodal probability measure. We show that this method can be seen as a variant of the well-known Wang–Landau algorithm Wang and Landau (Phys Rev E 64:056101, 2001a; Phys Rev Lett 86(10):2050–2053, 2001b). Adapting results on the convergence of the Wang-Landau algorithm obtained in Fort et al. (Math Comput 84(295):2297–2327, 2014a), we prove the convergence of the SHUS algorithm. We also compare the two methods in terms of efficiency. We finally propose a modification of the SHUS algorithm in order to increase its efficiency, and exhibit some similarities of SHUS with the well-tempered metadynamics method Barducci et al. (Phys Rev Lett 100:020,603, 2008).     

Estimation of the Hurst parameter in some fractional processes

We propose to estimate the Hurst parameter involved in fractional processes via a method based on the Karhunen–Loève expansion of a Gaussian process. We specifically investigate the cases of the fractional Brownian motion, the fractional Ornstein–Uhlenbeck family and the fractional Brownian bridge. We numerically compare our results with the ones obtained by the maximum-likelihood method, which show the validity of our proposal.     

Local linear estimation of a generalized regression function with functional dependent data

This work deals with a local linear non parametric estimation of the generalized regression function in the case of a scalar response variable given a random variable taking values in a semimetric space. The rates of pointwise and uniform almost complete convergence are established for the studied estimator when the sample is an α-mixing sequence. Two real datasets are used to illustrate the performance of a studied estimator with respect to the kernel method.     

Housing and building materials in low-income settlements in Dar es Salaam

Dar es Salaam (Tanzania) is one of the fastest growing cities in sub-Saharan Africa. As in other sub-Saharan cities, government housing programmes have reached only a tiny percentage of urban residents. The vast majority either build for themselves or live as tenants in swahili houses provided by others. This paper examines the process of house building in Dar es Salaam through a survey of 90 house-owners in nine low-income settlements in the city. It is shown that in the past decade the price of manufactured building materials has risen steeply following the implementation of liberalisation policies. Despite this, house owners still strive to build modern houses in concrete blocks, which offer higher standards of comfort and security as well as bringing in higher income from rents. Those who cannot afford to do so, are obliged either to rent one or more rooms in a house constructed by others or to build a modest house using traditional materials. Although these materials are available in the market, some housebuilders still collect their own. There is evidence of increasing distance from the source, the substitution of inferior species and declining quality, all of which indicate that resources are coming under pressure.     

Shelter Provision in the Context of Deforestation: Some Evidence From Mali

Deforestation in many parts of the developing world has to led to a deterioration in both the quantity and the quality of wood (timber) available for housebuilding, threatening the viability of traditional building techniques. This paper examines the effects of a deteriorating resource base upon traditional house building practices in Mopti (Mali), where the most common traditional house type requires large quantities of strong and durable timber for the construction of the roof. It is shown that the favoured “rônier” wood is no longer accessible to the poorer sections of the population, who have no alternative but to use lower quality wood and less durable species. This in turn means more frequent replacement and maintenance, thus putting greater pressure on the available natural resources, and on household budgets. Those who cannot afford to maintain their houses, live in a state of insecurity due to the fear of collapse from poor construction. The implications for policy are explored, both in Mali and in the wider context. Options include support for the introduction of new building techniques and for the planting of trees which provide good timber for building.     

Estimation of the Sobol indices in a linear functional multidimensional model

We consider a functional linear model where the explicative variables are known stochastic processes taking values in a Hilbert space, the main example is given by Gaussian processes in L²([0,1])${\mathbb{L}}^2([0,1])$. We propose estimators of the Sobol indices in this functional linear model. Our estimators are based on U-statistics. We prove the asymptotic normality and the efficiency of our estimators and we compare them from a theoretical and practical point of view with classical estimators of Sobol indices.     

Stochastic proximal-gradient algorithms for penalized mixed models

Motivated by penalized likelihood maximization in complex models, we study optimization problems where neither the function to optimize nor its gradient has an explicit expression, but its gradient can be approximated by a Monte Carlo technique. We propose a new algorithm based on a stochastic approximation of the proximal-gradient (PG) algorithm. This new algorithm, named stochastic approximation PG (SAPG) is the combination of a stochastic gradient descent step which—roughly speaking—computes a smoothed approximation of the gradient along the iterations, and a proximal step. The choice of the step size and of the Monte Carlo batch size for the stochastic gradient descent step in SAPG is discussed. Our convergence results cover the cases of biased and unbiased Monte Carlo approximations. While the convergence analysis of some classical Monte Carlo approximation of the gradient is already addressed in the literature (see Atchadé et al. in J Mach Learn Res 18(10):1–33, 2017), the convergence analysis of SAPG is new. Practical implementation is discussed, and guidelines to tune the algorithm are given. The two algorithms are compared on a linear mixed effect model as a toy example. A more challenging application is proposed on nonlinear mixed effect models in high dimension with a pharmacokinetic data set including genomic covariates. To our best knowledge, our work provides the first convergence result of a numerical method designed to solve penalized maximum likelihood in a nonlinear mixed effect model.

     

The effect of masks on the visual preference for faces in the first year of life

To prevent the spread of COVID-19, face masks were mandatory in many public spaces around the world. Since faces are the gateway to early social cognition, this raised major concerns about the effect face masks may have on infants' attention to faces as well as on their language and social development. The goal of the present study was to assess how face masks modulate infants' attention to faces over the course of the first year of life. We measured 3, 6, 9, and 12-month-olds’ looking behavior using a paired visual preference paradigm under two experimental conditions. First, we tested infants' preference for upright masked or unmasked faces of the same female individual. We found that regardless of age, infants looked equally long at the masked and unmasked faces. Second, we compared infants' attention to an upright masked versus an inverted masked face. Three- and 6-month-olds looked equally long to the masked faces when they were upright or inverted. However, 9- and 12-month-old infants showed a novelty preference for the inverted masked face. Our findings suggest that more experience with faces, including masked faces, leads to efficient adaptations of infants' visual system for processing impoverished social stimuli, such as partially occluded faces.     

Language familiarity influences own-race face recognition in 9- and 12-month-old infants

During their first year, infants attune to the faces and language(s) that are frequent in their environment. The present study investigates the impact of language familiarity on how French-learning 9- and 12-month-olds recognize own-race faces. In Experiment 1, infants were familiarized with the talking face of a Caucasian bilingual German-French speaker reciting a nursery rhyme in French (native condition) or in German (non-native condition). In the test phase, infants’ face recognition was tested by presenting a picture of the speaker's face they were familiarized with, side by side with a novel face. At 9 and 12 months, neither infants in the native condition nor the ones in the non-native condition clearly recognized the speaker's face. In Experiment 2, we familiarized infants with the still picture of the speaker's face, along with the auditory speech stream. This time, both 9- and 12-month-olds recognized the face of the speaker they had been familiarized with, but only if she spoke in their native language. This study shows that at least from 9 months of age, language modulates the way faces are recognized.     

WINNING IN PROFESSIONAL TEAM SPORTS: HISTORICAL MOMENTS

Our aims in this paper are to (1) examine the higher moments of the distribution of winning percentages and (2) discover economic implications of such an examination. The results prove useful to both current sports league policy questions and future research. We speculate that the institutional differences between North American pro leagues and European soccer leagues will prove fruitful areas for future research on the determination of competitive balance. (JEL C1, L83, Z20)