|
|
|
|
|
|
|
|
|
Bayesian learning
| Глоссарий по искусственному интеллекту |
Classical modeling methods usually produce a single model with fixed parameters. bayesian models instead
|
|
Bayesian, английский
Bayesian approach, английский
Байесовский подход
Bayesian belief function, английский
A belief function that corresponds to an ordinary probability function is referred to as a bayesian belief function. in this case, all of the probability mass is assigned to singleton sets, and none is assigned directly to unions of the elements. see also: belief function.
Bayesian chain, английский
Байесова цепь
Bayesian graphical model, английский
A probabilistic method of modeling for risk analysis to measure and reduce uncertainty.
Bayesian hierarchical model, английский
Bayesian hierarchical models specify layers of uncertainty on the phenomena being modeled and allow for multi-level heterogeneity in models for attributes. a base model is specified for the lowest level observations, and its parameters are specified by prior distributions for the parameters. each level above this also has a model that can include other parameters or prior distributions.
Bayesian knowledge discover, английский
Bayesian knowledge discoverer is a freely available program to construct and estimate bayesian belief networks. it can automatically estimate the network and export the results in the bayesian network interchange format (bnif). see also: bayesian network interchange format, belief net, http://kmi.open.ac.uk/projects/bkd
Bayesian methods, английский
Bayesian methods provide a formal method for reasoning about uncertain events. they are grounded in probability theory and use probabilistic techniques to assess and propagate the uncertainty. see also: certainty, fuzzy sets, possibility theory, probability. bayesian network (bn) a bayesian network is a graphical model that is used to represent probabilistic relationships among a set of attributes. the nodes, representing the state of attributes, are connected in a directed acyclic graph (dag). the arcs in the network represent probability models connecting the attributes. the probability models offer a flexible means to represent uncertainty in knowledge systems. they allow the system to specify the state of a set of attributes and infer the resulting distributions in the remaining attributes. the networks are called bayesian because they use the bayes theorem to propagate uncertainty throughout the network. note that the arcs are not required to represent causal directions but rather represent directions that probability propagates. see also: bayes theorem, belief net, influence diagrams. bayesian network interchange format (bnif) the bayesian network interchange format (bnif) is a proposed format for describing and interchanging belief networks. this will allow the sharing of knowledge bases that are represented as a bayesian network (bn) and allow the many bayes networks to interoperate. see also: bayesian network.
Bayesian network, английский
Вероятностная модель, представляющая набор переменных и их условных зависимостей через направленный ациклический граф.
Bayesian processing, английский
A form of inferencing in which the learner makes and discards hypotheses as the data get richer.
Bayesian statistics, английский
Байесовская статистика; байесова статистика bayonetбайонет
Bayesiano, испанский
Learn, английский
Learn (learnt, learned), английский
Learn addresses table, английский
Таблица пополняемых адресов
Learn mode, английский
Режим обучения
Learned, английский
Learned behavior, английский
Conducta aprendida
Learned words, английский
- serve to satisfy communicative demands of official, scientific, high poetry and poetic messages, authorial speech of creative prose; - mainly observed in the written form; - contribute to the message the tone of solemnity, sophistica
Learner, английский
N учащийся, обучающийся dictionary
Parameters, английский
Параметры
|
Monotone logic, английский
Classical logic is mostly monotone logic which has the property that if a=>b, the (a&c) also must =>b. this prevents a system based in monotone logic from retracting a conclusion based on new or conflicting evidence (propositions). see also: nonmonotone logic.
Local regression, английский
Classical regression approaches assume that a particular structural relationship holds over the entire range of predictors. local regression techniques assume that (usually) simple structures hold around each case in the data space, but that data points` "interestingness" decline the further one goes from its area. local regressions can typically reproduce a classical regression approach when the global regularity holds and outperform it when the regularity does not hold. smoothing splines, neural networks and regression and classification trees are all forms of local regression. see also: machine learning. locally optimal searches (solutions) a locally optimal search algorithm will use an algorithm that can find the best one-step or two-step result but is not guaranteed to find the globally optimum solution. often improved results can be found by iterating the search or by adding random perturbations to a local solutions.
|
|
|
|
|
|
|
