|
|
|
|
|
|
|
|
|
Bnif
| Глоссарий по искусственному интеллекту |
See: bayesian network interchange format.
|
|
Interchange, английский
- Пересечение дорог в разных уровнях, транспортная развязка
- A collection of one or more document instances that comprises a single transmission and is exchanged from application to application within an organization or from one trading partner to another.
- Операции (банкоматов) с кредитными карточками чужих
- A station where passenger may alight from one train and board another on a different route. in the us it is referred to as a "transfer".
|
Bayes classifier, английский
See: bayes rule.
Evidence theory, английский
See: belief functions, dempster-shaefer theory. evolutionary algorithm (ea) an evolutionary algorithm (ea) is a general class of fitting or maximization techniques. they all maintain a pool of structures or models that can be mutated and evolve. at every stage in the algorithm, each model is graded and the better models are allowed to reproduce or mutate for the next round. some techniques allow the successful models to crossbreed. they are all motivated by the biologic process of evolution. some techniques are asexual (so, there is no crosspage 99 breeding between techniques) while others are bisexual, allowing successful models to swap ``genetic" information. the asexual models allow a wide variety of different models to compete, while sexual methods require that the models share a common "genetic" code. evolutionary programming (ep) evolutionary programming (ep) is a machine learning (model estimation) technique that generalizes the genetic algorithm (ga) by keeping the evolutionary behavior and dropping the tight link to biologic genetics. it is essentially similar to asexual reproduction, in that each generation is evaluated and the most fit are most likely to reproduce. during reproduction, each "child" model is allowed to mutate. this method focuses less on the internal representation of the model and more on the behavior of the models. it becomes easier to mix diverse types of models in the same mix. note that this technique usually selects the "winners" stochastically rather than deterministically.
|
|
|
|
|
|
|
