Mémoire de travailLe système cognitif fonctionne en acquérant, filtrant et traitant des informations vitales, utiles, potentiellement utiles à court, moyen et long termes ; il a donc besoin de stocker (mémoriser) ces informations. Le cerveau semble pour cela disposer de systèmes différents, mais complémentaires, de mémoire à long terme et de mémoire à court terme. La notion de mémoire de travail, apparue dans les années 1970 désigne .
Statistical learning in language acquisitionStatistical learning is the ability for humans and other animals to extract statistical regularities from the world around them to learn about the environment. Although statistical learning is now thought to be a generalized learning mechanism, the phenomenon was first identified in human infant language acquisition. The earliest evidence for these statistical learning abilities comes from a study by Jenny Saffran, Richard Aslin, and Elissa Newport, in which 8-month-old infants were presented with nonsense streams of monotone speech.
Sciences numériquesLes sciences numériques (traduction de l'anglais computational sciences), autrement dénommées calcul scientifique ou informatique scientifique, ont pour objet la construction de modèles mathématiques et de méthodes d'analyse quantitative, en se basant sur l'utilisation des sciences du numérique, pour analyser et résoudre des problèmes scientifiques. Cette approche scientifique basée sur un recours massif aux modélisations informatiques et mathématiques et à la simulation se décline en : médecine numérique, biologie numérique, archéologie numérique, mécanique numérique, par exemple.
Computational anatomyComputational anatomy is an interdisciplinary field of biology focused on quantitative investigation and modelling of anatomical shapes variability. It involves the development and application of mathematical, statistical and data-analytical methods for modelling and simulation of biological structures. The field is broadly defined and includes foundations in anatomy, applied mathematics and pure mathematics, machine learning, computational mechanics, computational science, biological imaging, neuroscience, physics, probability, and statistics; it also has strong connections with fluid mechanics and geometric mechanics.
Mémoire temporelle et hiérarchiqueLa mémoire temporelle et hiérarchique (en anglais Hierarchical temporal memory (HTM)) est un modèle d'apprentissage automatique développé par Jeff Hawkins et Dileep George de la compagnie Numenta. Il modélise certaines propriétés structurelles et algorithmiques du néocortex. C'est un modèle biomimétique fondé sur le paradigme mémoire-prédiction, une théorie du fonctionnement du cerveau élaborée par Jeff Hawkins dans son livre On Intelligence.
Computational engineeringComputational Engineering is an emerging discipline that deals with the development and application of computational models for engineering, known as Computational Engineering Models or CEM. At this time, various different approaches are summarized under the term Computational Engineering, including using computational geometry and virtual design for engineering tasks, often coupled with a simulation-driven approach In Computational Engineering, algorithms solve mathematical and logical models that describe engineering challenges, sometimes coupled with some aspect of AI, specifically Reinforcement Learning.
Inverse probabilityIn probability theory, inverse probability is an obsolete term for the probability distribution of an unobserved variable. Today, the problem of determining an unobserved variable (by whatever method) is called inferential statistics, the method of inverse probability (assigning a probability distribution to an unobserved variable) is called Bayesian probability, the "distribution" of data given the unobserved variable is rather the likelihood function (which is not a probability distribution), and the distribution of an unobserved variable, given both data and a prior distribution, is the posterior distribution.
Two-streams hypothesisThe two-streams hypothesis is a model of the neural processing of vision as well as hearing. The hypothesis, given its initial characterisation in a paper by David Milner and Melvyn A. Goodale in 1992, argues that humans possess two distinct visual systems. Recently there seems to be evidence of two distinct auditory systems as well. As visual information exits the occipital lobe, and as sound leaves the phonological network, it follows two main pathways, or "streams".
Time delay neural networkTime delay neural network (TDNN) is a multilayer artificial neural network architecture whose purpose is to 1) classify patterns with shift-invariance, and 2) model context at each layer of the network. Shift-invariant classification means that the classifier does not require explicit segmentation prior to classification. For the classification of a temporal pattern (such as speech), the TDNN thus avoids having to determine the beginning and end points of sounds before classifying them.
Loss functions for classificationIn machine learning and mathematical optimization, loss functions for classification are computationally feasible loss functions representing the price paid for inaccuracy of predictions in classification problems (problems of identifying which category a particular observation belongs to). Given as the space of all possible inputs (usually ), and as the set of labels (possible outputs), a typical goal of classification algorithms is to find a function which best predicts a label for a given input .
Algorithme de rechercheEn informatique, un algorithme de recherche est un type d'algorithme qui, pour un domaine, un problème de ce domaine et des critères donnés, retourne en résultat un ensemble de solutions répondant au problème. Supposons que l'ensemble de ses entrées soit divisible en sous-ensemble, par rapport à un critère donné, qui peut être, par exemple, une relation d'ordre. De façon générale, un tel algorithme vérifie un certain nombre de ces entrées et retourne en sortie une ou plusieurs des entrées visées.
Codage neuronalLe codage neuronal désigne, en neurosciences, la relation hypothétique entre le stimulus et les réponses neuronales individuelles ou globales. C'est une théorie sur l'activité électrique du système nerveux, selon laquelle les informations, par exemple sensorielles, numériques ou analogiques, sont représentées dans le cerveau par des réseaux de neurones. Le codage neuronal est lié aux concepts du souvenir, de l'association et de la mémoire sensorielle.
