Statistical learning theory for geospatial data. Case study: Aral sea

In recent years there has been an explosive growth in the development of adaptive and data driven methods. One of the efficient and data-driven approaches is based on statistical learning theory (Vapnik 1998). The theory is based on Structural Risk Minimisation (SRM) principle and has a solid statistical background. When applying SRM we are trying not only to reduce training error ? to fit the available data with a model, but also to reduce the complexity of the model and to reduce generalisation error. Many nonlinear learning procedures recently developed in neural networks and statistics can be understood and interpreted in terms of the structural risk minimisation inductive principle. A recent methodology based on SRM is called Support Vector Machines (SVM). At present SLT is still under intensive development and SVM find new areas of application (www.kernel-machines.org). SVM develop robust and non linear data models with excellent generalisation abilities that is very important both for monitoring and forecasting. SVM are extremely good when input space is high dimensional and training data set i not big enough to develop corresponding nonlinear model. Moreover, SVM use only support vectors to derive decision boundaries. It opens a way to sampling optimization, estimation of noise in data, quantification of data redundancy etc. Presentation of SVM for spatially distributed data is given in (Kanevski and Maignan 2004).

Active learning methods for remote sensing image classification

In this paper, we propose two active learning algorithms for semiautomatic definition of training samples in remote sensing image classification. Based on predefined heuristics, the classifier ranks the unlabeled pixels and automatically chooses those that are considered the most valuable for its improvement. Once the pixels have been selected, the analyst labels them manually and the process is iterated. Starting with a small and nonoptimal training set, the model itself builds the optimal set of samples which minimizes the classification error. We have applied the proposed algorithms to a variety of remote sensing data, including very high resolution and hyperspectral images, using support vector machines. Experimental results confirm the consistency of the methods. The required number of training samples can be reduced to 10% using the methods proposed, reaching the same level of accuracy as larger data sets. A comparison with a state-of-the-art active learning method, margin sampling, is provided, highlighting advantages of the methods proposed. The effect of spatial resolution and separability of the classes on the quality of the selection of pixels is also discussed.

Intrahippocampal cholinergic grafts in aged rats compensate impairments in a radial maze and in a place learning task

Age-related cognitive impairments were studied in rats kept in semi-enriched conditions during their whole life, and tested during ontogeny and adult life in various classical spatial tasks. In addition, the effect of intrahippocampal grafts of fetal septal-diagonal band tissue, rich in cholinergic neurons, was studied in some of these subjects. The rats received bilateral cell suspensions when aged 23-24 months. Starting 4 weeks after grafting, they were trained during 5 weeks in an 8-arm maze made of connected plexiglass tunnels. No age-related impairment was detected during the first eight trials, when the maze shape was that of a classical radial maze in which the rats had already been trained when young. The older rats were impaired when the task was made more difficult by rendering two arms parallel to each other. They developed an important neglect of one of the parallel tunnels resulting in a high amount of errors before completion of the task. In addition, the old rats developed a systematic response pattern of visits to adjacent arms in a sequence, which was not observed in the younger subjects. None of these behaviours were observed in the old rats with a septal transplant. Sixteen weeks after grafting, another experiment was conducted in a homing hole board task. Rats were allowed to escape from a large circular arena through one hole out of many, and to reach home via a flexible tube under the table. The escape hole was at a fixed position according to distant room cues, and olfactory cues were made irrelevant by rotating the table between the trials. An additional cue was placed on the escape position. No age-related difference in escape was observed during training. During a probe trial with no hole connected and no proximal cue present, the old untreated rats were less clearly focussed on the training sector than were either the younger or the grafted old subjects. Taken together, these experiments indicate that enriched housing conditions and spatial training during adult life do not protect against all age-related deterioration in spatial ability. However, it might be that the considerable improvement observed in the grafted subjects results from an interaction between the graft treatment and the housing conditions.

Effects of inbreeding on aversive learning in Drosophila.

Inbreeding adversely affects life history traits as well as various other fitness-related traits, but its effect on cognitive traits remains largely unexplored, despite their importance to fitness of many animals under natural conditions. We studied the effects of inbreeding on aversive learning (avoidance of an odour previously associated with mechanical shock) in multiple inbred lines of Drosophila melanogaster derived from a natural population through up to 12 generations of sib mating. Whereas the strongly inbred lines after 12 generations of inbreeding (0.75<F<0.93) consistently showed reduced egg-to-adult viability (on average by 28%), the reduction in learning performance varied among assays (average=18% reduction), being most pronounced for intermediate conditioning intensity. Furthermore, moderately inbred lines (F=0.38) showed no detectable decline in learning performance, but still had reduced egg-to-adult viability, which indicates that overall inbreeding effects on learning are mild. Learning performance varied among strongly inbred lines, indicating the presence of segregating variance for learning in the base population. However, the learning performance of some inbred lines matched that of outbred flies, supporting the dominance rather than the overdominance model of inbreeding depression for this trait. Across the inbred lines, learning performance was positively correlated with the egg-to-adult viability. This positive genetic correlation contradicts a trade-off observed in previous selection experiments and suggests that much of the genetic variation for learning is owing to pleiotropic effects of genes affecting functions related to survival. These results suggest that genetic variation that affects learning specifically (rather than pleiotropically through general physiological condition) is either low or mostly due to alleles with additive (semi-dominant) effects.

Multisensory Integration and Perceptual Enhancement

We perceive our environment through multiple sensory channels. Nonetheless, research has traditionally focused on the investigation of sensory processing within single modalities. Thus, investigating how our brain integrates multisensory information is of crucial importance for understanding how organisms cope with a constantly changing and dynamic environment. During my thesis I have investigated how multisensory events impact our perception and brain responses, either when auditory-visual stimuli were presented simultaneously or how multisensory events at one point in time impact later unisensory processing. In "Looming signals reveal synergistic principles of multisensory integration" (Cappe, Thelen et al., 2012) we investigated the neuronal substrates involved in motion detection in depth under multisensory vs. unisensory conditions. We have shown that congruent auditory-visual looming (i.e. approaching) signals are preferentially integrated by the brain. Further, we show that early effects under these conditions are relevant for behavior, effectively speeding up responses to these combined stimulus presentations. In "Electrical neuroimaging of memory discrimination based on single-trial multisensory learning" (Thelen et al., 2012), we investigated the behavioral impact of single encounters with meaningless auditory-visual object parings upon subsequent visual object recognition. In addition to showing that these encounters lead to impaired recognition accuracy upon repeated visual presentations, we have shown that the brain discriminates images as soon as ~100ms post-stimulus onset according to the initial encounter context. In "Single-trial multisensory memories affect later visual and auditory object recognition" (Thelen et al., in review) we have addressed whether auditory object recognition is affected by single-trial multisensory memories, and whether recognition accuracy of sounds was similarly affected by the initial encounter context as visual objects. We found that this is in fact the case. We propose that a common underlying brain network is differentially involved during encoding and retrieval of images and sounds based on our behavioral findings. - Nous percevons l'environnement qui nous entoure à l'aide de plusieurs organes sensoriels. Antérieurement, la recherche sur la perception s'est focalisée sur l'étude des systèmes sensoriels indépendamment les uns des autres. Cependant, l'étude des processus cérébraux qui soutiennent l'intégration de l'information multisensorielle est d'une importance cruciale pour comprendre comment notre cerveau travail en réponse à un monde dynamique en perpétuel changement. Pendant ma thèse, j'ai ainsi étudié comment des événements multisensoriels impactent notre perception immédiate et/ou ultérieure et comment ils sont traités par notre cerveau. Dans l'étude " Looming signals reveal synergistic principles of multisensory integration" (Cappe, Thelen et al., 2012), nous nous sommes intéressés aux processus neuronaux impliqués dans la détection de mouvements à l'aide de l'utilisation de stimuli audio-visuels seuls ou combinés. Nos résultats ont montré que notre cerveau intègre de manière préférentielle des stimuli audio-visuels combinés s'approchant de l'observateur. De plus, nous avons montré que des effets précoces, observés au niveau de la réponse cérébrale, influencent notre comportement, en accélérant la détection de ces stimuli. Dans l'étude "Electrical neuroimaging of memory discrimination based on single-trial multisensory learning" (Thelen et al., 2012), nous nous sommes intéressés à l'impact qu'a la présentation d'un stimulus audio-visuel sur l'exactitude de reconnaissance d'une image. Nous avons étudié comment la présentation d'une combinaison audio-visuelle sans signification, impacte, au niveau comportementale et cérébral, sur la reconnaissance ultérieure de l'image. Les résultats ont montré que l'exactitude de la reconnaissance d'images, présentées dans le passé, avec un son sans signification, est inférieure à celle obtenue dans le cas d'images présentées seules. De plus, notre cerveau différencie ces deux types de stimuli très tôt dans le traitement d'images. Dans l'étude "Single-trial multisensory memories affect later visual and auditory object recognition" (Thelen et al., in review), nous nous sommes posés la question si l'exactitude de ia reconnaissance de sons était affectée de manière semblable par la présentation d'événements multisensoriels passés. Ceci a été vérifié par nos résultats. Nous avons proposé que cette similitude puisse être expliquée par le recrutement différentiel d'un réseau neuronal commun.

L'interdiction de discrimination en raison de la nationalité au sens de l'Accord sur la libre circulation des personnes

Introduction : 6 décembre 1992: le peuple suisse rejette l'Accord sur l'Espace économique européen (EEE). Le Conseil fédéral décide alors de privilégier la voie bilatérale pour réglementer ses relations avec l'Union européenne (UE). Près de sept ans plus tard, le 21 juin 1999, un premier paquet de sept accords bilatéraux sont signés à Luxembourg. L'Accord sur la libre circulation des personnes (ALCP ou Accord) en constitue certainement « la partie la plus importante ». L'ALCP a pour objectif de garantir la libre circulation des travailleurs salariés, des travailleurs indépendants ainsi que de certaines catégories de non-actifs, et de faciliter la prestation de services sur le territoire des parties contractantes. Or la réalisation de ces objectifs dépend directement de la mise en oeuvre de l'interdiction de discrimination en raison de la nationalité. Au coeur même du principe de libre circulation, l'interdiction de discrimination en raison de la nationalité constitue, sans aucun doute, la clé de voûte de l'Accords. Partant de ce constat, il conviendra d'étudier dans un premier temps le principe de libre circulation (Partie I), puis de nous pencher dans un second temps sur l'interdiction de discrimination en raison de la nationalité (Partie II). La première partie de notre étude (consacrée à la libre circulation des personnes en vertu de l'ALCP) débutera par un survol des dispositions essentielles à l'appréhension d'un droit fondamental à la libre circulation, ainsi que par une présentation succincte du régime communautaire de libre circulation. Après avoir rappelé les étapes du cheminement ayant conduit à l'adoption de l'ALCP, et après avoir défini les différentes sources régissant la libre circulation entre la Suisse et les Etats membres de l'UE, nous analyserons ensuite l'ALCP lui-même, en procédant à la détermination de son champ d'application ainsi qu'à l'examen des principes généraux qui le régissent. Cet examen nous amènera alors à déterminer le rôle que devrait jouer la jurisprudence de la Cour de justice des Communautés européennes (CJCE) dans le cadre de l'Accord. Fort de cette analyse, nous conclurons cette première partie par une présentation du régime de libre circulation tel qu'institué par l'ALCP, sans oublier bien sûr d'évoquer, finalement, la portée du concept d'entrave et de la notion de citoyenneté européenne dans le cadre de ce régime. La seconde partie de notre étude (consacrée à l'interdiction de discrimination en raison de la nationalité selon l'ALCP) identifiera, à titre liminaire, les différentes dispositions de l'Accord visant à prohiber les discriminations en raison de la nationalité. Ces dispositions mises en lumière, nous détaillerons alors les étapes de l'examen permettant de relever la présence d'une discrimination en raison de la nationalité, en nous appuyant sur la jurisprudence de la Cour de justice et celle du Tribunal fédéral d'une part, et sur la doctrine développée en matière d'interdiction de discrimination d'autre part. Dans la foulée, nous pourrons ainsi présenter les différentes formes que peut revêtir une discrimination. Nous examinerons, en particulier, le régime prévu par la jurisprudence de la Cour de justice et celle du Tribunal fédéral en matière de discriminations à rebours présentant un élément transfrontalier. Suite de quoi nous serons en mesure - après examen des jurisprudences allemande et autrichienne relatives aux situations internes de discriminations à rebours - de procéder à une appréciation critique de la solution adoptée par le Tribunal fédéral en pareilles situations. A la suite de ces deux chapitres portant sur la notion et sur les formes de discriminations, il y aura lieu de définir le champ d'application matériel et personnel des différentes dispositions de l'Accord visant à interdire les discriminations en raison de la nationalité. Lors de la définition du champ d'application personnel en particulier, il conviendra, en premier lieu, d'identifier les personnes susceptibles d'invoquer ces différentes dispositions (les bénéficiaires). Cet examen débutera par l'identification des conditions générales communes à l'ensemble des dispositions étudiées. Il se poursuivra par l'analyse des conditions spécifiques de chaque norme, et se terminera par une présentation du régime particulier réservé aux membres de la famille et aux travailleurs détachés. En second lieu, il s'agira de déterminer les sujets de droit à l'encontre desquels ces dispositions sont opposables (les destinataires). Plus précisément, il s'agira d'examiner dans quelle mesure les particuliers, en sus des Etats, peuvent être destinataires des différentes obligations de l'ALCP en matière de libre circulation et d'interdiction de discrimination. Les champs d'application ayant été définis, il nous restera alors à examiner les différentes dispositions de l'Accord susceptibles de limiter la portée du principe de non-discrimination. Pour ce faire, nous analyserons en détail l'article 5 Annexe I-ALCP qui permet de limiter les droits consacrés par l'Accord pour des raisons d'ordre public, de sécurité publique ou de santé publique. Nous nous pencherons aussi sur les limitations liées à l'exercice de la puissance publique, consacrées par les articles 10, 16 et 22 al. 1 Annexe I-ALCP. Après un passage en revue des dernières dispositions de l'Accord prévoyant des limitations, nous examinerons ensuite dans quelle mesure certains principes non expressément prévus par l'Accord sont également susceptibles de justifier une différence de traitement en raison de la nationalité. Nous conclurons cette étude en analysant les implications juridiques d'une violation de l'interdiction de discrimination, ceci non seulement dans les relations liant l'Etat à un particulier, mais aussi dans celles liant deux particuliers entre eux. Dans le premier cas de figure nous verrons qu'il y a lieu de différencier l'analyse en fonction du type d'acte en question - les implications d'une violation n'étant pas les mêmes selon que l'on se trouve en présence d'une norme, d'une décision ou d'un contrat (de droit administratif ou de droit privé) présentant un caractère discriminatoire. Dans le second cas de figure, il s'agira cette fois-ci de distinguer les implications des conditions discriminatoires en fonction des différentes phases d'une relation contractuelle - aux stades de la conclusion d'un contrat, de son exécution et de sa résiliation.

Dynamic properties of human brain structure: learning-related changes in cortical areas and associated fiber connections.

Recent findings in neuroscience suggest that adult brain structure changes in response to environmental alterations and skill learning. Whereas much is known about structural changes after intensive practice for several months, little is known about the effects of single practice sessions on macroscopic brain structure and about progressive (dynamic) morphological alterations relative to improved task proficiency during learning for several weeks. Using T1-weighted and diffusion tensor imaging in humans, we demonstrate significant gray matter volume increases in frontal and parietal brain areas following only two sessions of practice in a complex whole-body balancing task. Gray matter volume increase in the prefrontal cortex correlated positively with subject's performance improvements during a 6 week learning period. Furthermore, we found that microstructural changes of fractional anisotropy in corresponding white matter regions followed the same temporal dynamic in relation to task performance. The results make clear how marginal alterations in our ever changing environment affect adult brain structure and elucidate the interrelated reorganization in cortical areas and associated fiber connections in correlation with improvements in task performance.

Semantic domains and supersens tagging for domain-specific ontology learning

In this paper we propose a novel unsupervised approach to learning domain-specific ontologies from large open-domain text collections. The method is based on the joint exploitation of Semantic Domains and Super Sense Tagging for Information Retrieval tasks. Our approach is able to retrieve domain specific terms and concepts while associating them with a set of high level ontological types, named supersenses, providing flat ontologies characterized by very high accuracy and pertinence to the domain.

Analysis, modelling and classification of geospatial data using machine learning

The research considers the problem of spatial data classification using machine learning algorithms: probabilistic neural networks (PNN) and support vector machines (SVM). As a benchmark model simple k-nearest neighbor algorithm is considered. PNN is a neural network reformulation of well known nonparametric principles of probability density modeling using kernel density estimator and Bayesian optimal or maximum a posteriori decision rules. PNN is well suited to problems where not only predictions but also quantification of accuracy and integration of prior information are necessary. An important property of PNN is that they can be easily used in decision support systems dealing with problems of automatic classification. Support vector machine is an implementation of the principles of statistical learning theory for the classification tasks. Recently they were successfully applied for different environmental topics: classification of soil types and hydro-geological units, optimization of monitoring networks, susceptibility mapping of natural hazards. In the present paper both simulated and real data case studies (low and high dimensional) are considered. The main attention is paid to the detection and learning of spatial patterns by the algorithms applied.

Attention-related potentials allow for a highly accurate discrimination of mild cognitive impairment subtypes.

The three most frequent forms of mild cognitive impairment (MCI) are single-domain amnestic MCI (sd-aMCI), single-domain dysexecutive MCI (sd-dMCI) and multiple-domain amnestic MCI (md-aMCI). Brain imaging differences among single domain subgroups of MCI were recently reported supporting the idea that electroencephalography (EEG) functional hallmarks can be used to differentiate these subgroups. We performed event-related potential (ERP) measures and independent component analysis in 18 sd-aMCI, 13 sd-dMCI and 35 md-aMCI cases during the successful performance of the Attentional Network Test. Sensitivity and specificity analyses of ERP for the discrimination of MCI subgroups were also made. In center-cue and spatial-cue warning stimuli, contingent negative variation (CNV) was elicited in all MCI subgroups. Two independent components (ICA1 and 2) were superimposed in the time range on the CNV. The ICA2 was strongly reduced in sd-dMCI compared to sd-aMCI and md-aMCI (4.3 vs. 7.5% and 10.9% of the CNV component). The parietal P300 ERP latency increased significantly in sd-dMCI compared to md-aMCI and sd-aMCI for both congruent and incongruent conditions. This latency for incongruent targets allowed for a highly accurate separation of sd-dMCI from both sd-aMCI and md-aMCI with correct classification rates of 90 and 81%, respectively. This EEG parameter alone performed much better than neuropsychological testing in distinguishing sd-dMCI from md-aMCI. Our data reveal qualitative changes in the composition of the neural generators of CNV in sd-dMCI. In addition, they document an increased latency of the executive P300 component that may represent a highly accurate hallmark for the discrimination of this MCI subgroup in routine clinical settings.