DataFranca - Contributions [fr]

Arbre de décision

2020-01-19T13:31:48Z

Gdpelletier : français modifié; anglais ajouté

[[Catégorie:Intelligence artificielle]]
[[Catégorie:GRAND LEXIQUE FRANÇAIS]]


==Définition==
Outil d'aide à la décision sous la forme graphique d'un arbre avec sa racine en haut; les différentes décisions possibles étant situées aux extrémités des branches et sont adoptées en fonction de la décision prise à chaque étape. Cette méthodologie est plus communément appelée arbre de classification à partir des données. Les arbres de régression sont représentés de la même manière, sauf qu’ils prédisent des valeurs continues comme le prix d'une maison.

==Français==
'''arbre de décision''' n.m.

==Anglais==
'''decision tree'''

''A decision tree is drawn upside down with its root at the top. This methodology is more commonly known as '''learning decision tree from data''' and is called '''Classification tree.''' '''Regression trees''' are represented in the same manner, just they predict continuous values like price of a house. In general, Decision Tree algorithms are referred to as CART or Classification and Regression Trees.''

source:https://towardsdatascience.com/decision-trees-in-machine-learning-641b9c4e8052



[https://fr.wikipedia.org/wiki/Arbre_de_d%C3%A9cision ''Source: Wikipedia'']

Analyse factorielle

2020-01-18T16:51:12Z

Gdpelletier : définition anglaise

==en construction==
[[Catégorie:Vocabulaire]]
[[Catégorie:App-profond-livre]]
[[Catégorie:Apprentissage profond]]
[[Catégorie:Scotty]]

==Définition==
L'analyse factorielle est une méthode statistique utilisée pour décrire la variabilité entre les variables corrélées observées en termes d'un nombre potentiellement inférieur de variables non observées appelées facteurs. Par exemple, il est possible que les variations de six variables observées reflètent principalement les variations de deux variables (sous-jacentes) non observées.

L'analyse factorielle recherche de telles variations conjointes en réponse à des variables latentes non observées. Les variables observées sont modélisées sous forme de combinaisons linéaires des facteurs potentiels, plus des termes "erreur". L'analyse factorielle vise à trouver des variables latentes indépendantes.

==Français==
'''Analyse factorielle'''

==Anglais==
'''Factor analysis'''

Factor analysis is a method for modeling observed variables and their covariance structure in terms of unobserved variables (i.e., factors). There are two types of factor analyses, exploratory and confirmatory. Exploratory factor analysis (EFA) is method to explore the underlying structure of a set of observed variables, and is a crucial step in the scale development process. The first step in EFA is factor extraction. Common factor analysis models can be estimated using various estimation methods such as principal axis factoring and maximum likelihood, and we will compare the practical differences between these two methods. After extracting the best factor structure, we can obtain a more interpretable factor solution through factor rotation. Confirmatory factor analysis (CFA), which is a method to verify a factor structure that has already been defined. Topics to discuss include identification, model fit, and examples of a one-factor, uncorrelated two-factor and correlated two-factor model.

Source: https://stats.idre.ucla.edu/spss/seminars/introduction-to-factor-analysis/



[https://fr.wikipedia.org/wiki/Analyse_factorielle Source : wikipedia]

[https://www.apprentissageprofond.org/ Source : ''L'apprentissage profond'', Ian Goodfellow, Yoshua Bengio et Aaron Courville Éd. Massot 2018 page 488]

Analyse des caractéristiques lentes

2020-01-18T16:35:50Z

Gdpelletier : ajout anglais

==en construction==
[[Catégorie:Vocabulaire]]
[[Catégorie:App-profond-livre]]
[[Catégorie:Apprentissage profond]]
[[Catégorie:Scotty]]

==Définition==
L'analyse des caractéristiques lentes est un modèle linéaire factoriel qui utilise l'information de signaux temporels pour apprendre des caractéristiques invariantes ( Wiskott et Sejnowski, 2002 )

==Français==
'''Analyse des caractéristiques lentes'''

==Anglais==
'''Slow feature analysis '''

'''Slow feature analysis (SFA)''' is an unsupervised learning algorithm for extracting slowly varying features from a quickly varying input signal. It has been successfully applied, e.g., to the self-organization of complex-cell receptive fields, the recognition of whole objects invariant to spatial transformations, the self-organization of place-cells, extraction of driving forces, and to nonlinear blind source separation.

[http://www.scholarpedia.org/article/Slow_feature_analysis Source: http://www.scholarpedia.org/article/Slow_feature_analysis]



[https://www.apprentissageprofond.org/ Source : ''L'apprentissage profond'', Ian Goodfellow, Yoshua Bengio et Aaron Courville Éd. Massot 2018 p.492]

Décideur

2019-03-08T11:07:17Z

Gdpelletier : Page créée avec « Decision-making skills have to be in place before a team can get value out of data. This individual is responsible for identifying decisions worth making with data, framin... »

Decision-making skills have to be in place before a team can get value out of data.
This individual is responsible for identifying decisions worth making with data, framing them (everything from designing metrics to calling the shots on statistical assumptions), and determining the required level of analytical rigor based on potential impact on the business. Look for a deep thinker who doesn’t keep saying, “Oh, whoops, that didn’t even occur to me as I was thinking through this decision.” They’ve already thought of it. And that. And that too.
file:///C:/Users/user/Documents/texte%20mars/Top%2010%20roles%20in%20AI%20and%20data%20science%20–%20Hacker%20Noon.pdf

Apprentissage par renforcement avec borne de confiance supérieure

2019-03-08T10:49:52Z

Gdpelletier : Page créée avec « Recall the general setup for reinforcement learning: we have well-defined actions that we can take, so we let the machine figure out how to maximize its reward based on th... »

Recall the general setup for reinforcement learning: we have well-defined actions that we can take, so we let the machine figure out how to maximize its reward based on the consequences of those actions.

The Upper Confidence Bound algorithm is a formalization of this idea, where the machine attempts to determine a single action it can take that will maximize its expected return.

https://opendatascience.com/machine-learning-for-beginners/

Séparateur à vaste marge à noyau

2019-03-08T10:47:11Z

Gdpelletier : Page créée avec « The kernel support vector machine is essentially the same as the standard SVM, with a cool trick that allows it to discover non-linear decision boundaries. Instead of us... »

The kernel support vector machine is essentially the same as the standard SVM, with a cool trick that allows it to discover non-linear decision boundaries.

Instead of using the data as-is, we throw our data into something called a kernel. The kernel is any function that takes an input with a given dimensionality and returns an output with higher dimensionality, effectively adding more features to your examples.

Recherche de règles d'association

2019-03-08T10:21:48Z

Gdpelletier : Page créée avec « discovering rules for observing A given B. This is closely related to clustering in that we’re attempting to find connections between events. However, the difference is... »

discovering rules for observing A given B. This is closely related to clustering in that we’re attempting to find connections between events. However, the difference is the approach. Instead of drawing bounds on a region and seeing if someone fits in that bucket, we use the frequency of a collection of discrete observations to create priors: what’s the probability of observing A, or B, or C? From these, we figure out what the probability is of observing A given B, or observing C given A and B. This is just Bayesian statistics: P(A), P(B | A), P(C | A, B), and so forth.

Machine Learning for Beginners – a How-to Guide
https://opendatascience.com/machine-learning-for-beginners/

Programmation probabiliste

2019-02-05T19:44:23Z

Gdpelletier : AI Knowledge Map, Forbes

__NOTOC__
== Domaine ==
[[Category:Vocabulaire]]Vocabulaire 

 
 

== Définition ==
A framework that does not force you to hardcode specific variable but rather works with probabilistic models. Bayesian Program Synthesis (BPS) is somehow a form of probabilistic programming, where Bayesian programs write new Bayesian programs (instead of humans do it, as in the broader probabilistic programming approach)
 
 

== Français ==
'''Terme_français'''
 
 

== Anglais ==
'''Terme_Anglais'''

Système d'IA autonome

2019-02-05T19:33:27Z

Gdpelletier : AI Knowledge Map, Forbes

__NOTOC__
== Domaine ==
[[Category:Vocabulaire]]Vocabulaire 

 
 

== Définition ==
Sub-field that lies at the intersection between robotics and intelligent systems (e.g., intelligent perception, dexterous object manipulation, plan-based robot control, etc.)
 
 

== Français ==
'''Terme_français'''
 
 

== Anglais ==
'''Terme_Anglais'''

Apprentissage profond

2019-02-05T19:30:52Z

Gdpelletier : /* deep learning */

__NOTOC__
== Domaine ==
[[Category:intelligence artificielle]]intelligence artificielle 
[[Category:GDT]]GDT 
[[category:Apprentissage profond ]]Apprentissage profond
[[Category:GRAND LEXIQUE FRANÇAIS]]GRAND LEXIQUE FRANÇAIS 
 

== Définition ==
Mode d'apprentissage automatique généralement effectué par un réseau de neurones artificiels composé de plusieurs couches de neurones qui, en interagissant entre elles, permettent aux ordinateurs d'apprendre progressivement et efficacement à partir de mégadonnées.

Notes
L'apprentissage profond est inspiré des connaissances en neurosciences. Il facilite notamment l'acquisition de règles complexes et la perception des signaux (images, vidéos, sons, parole, etc.).
La recherche en apprentissage profond s'applique, entre autres, à la reconnaissance de la parole et à la reconnaissance des formes ainsi qu'à la robotique, à la vision par ordinateur et au traitement automatique des langues.
 
 

== Français ==
=== apprentissage profond n. m.===
=== apprentissage en profondeur n. m. ===
 
 

== Anglais ==
=== deep learning ===
(a neural net with multiple layers)

=== deep machine learning ===
=== deep structured learning ===
 
 

[http://gdt.oqlf.gouv.qc.ca/ficheOqlf.aspx?Id_Fiche=26532876 ''Source: Grand Dictionnaire Terminologique'' ]
 
 
[https://datafranca.org/lexique/apprentissage-profond/ ''Publié : datafranca.org'' ]

Automatisation robotisée de processus

2019-02-05T19:22:31Z

Gdpelletier : AI Knowledge Map, Forbes

__NOTOC__
== Domaine ==
[[Category:Vocabulaire]]Vocabulaire 

 
 

== Définition ==
Technology that extracts the list of rules and actions to perform by watching the user doing a certain task
 
 

== Français ==
'''Terme_français'''
 
 

== Anglais ==
'''Terme_Anglais'''

Approche statistique

2019-02-05T19:18:01Z

Gdpelletier : AI Knowledge Map, Forbes

__NOTOC__
== Domaine ==
[[Category:Vocabulaire]]Vocabulaire 

 
 

== Définition ==
Statistical Approach is based on mathematical tools to solve specific sub-problems
 
 

== Français ==
'''Terme_français'''
 
 

== Anglais ==
'''Terme_Anglais'''

Intelligence incorporée

2019-02-05T19:08:54Z

Gdpelletier : Page créée avec « __NOTOC__ == Domaine == Category:VocabulaireVocabulaire == Définition == Engineering toolbopx, which assumes that a body (or at leat a partial s... »

__NOTOC__
== Domaine ==
[[Category:Vocabulaire]]Vocabulaire 

 
 

== Définition ==
Engineering toolbopx, which assumes that a body (or at leat a partial set of functions such a movement, perception, interaction and visualisation) is required for higher intelligence.
 
 

== Français ==
'''Terme_français'''
 
 

== Anglais ==
'''Terme_Anglais'''

Méthode probabiliste

2019-02-05T19:03:36Z

Gdpelletier : Page créée avec « __NOTOC__ == Domaine == Category:VocabulaireVocabulaire == Définition == tools that allow agent to act in incomplete information scenarios <br /... »

__NOTOC__
== Domaine ==
[[Category:Vocabulaire]]Vocabulaire 

 
 

== Définition ==
tools that allow agent to act in incomplete information scenarios
 
 

== Français ==
'''Terme_français'''
 
 

== Anglais ==
'''Terme_Anglais'''

Deep learning

2018-04-02T16:20:33Z

Gdpelletier : Page créée avec « __NOTOC__ == Domaine == Category:VocabularyVocabulary == Définition == Deep learning is a subfield of machine learning concerned with algorithms insp... »

__NOTOC__

== Domaine ==
[[Category:Vocabulary]]Vocabulary 

 

== Définition ==
Deep learning is a subfield of machine learning concerned with algorithms inspired by the structure and function of the brain called artificial neural networks.”
— Jason Brownlee from Machine Learning Mastery
 
== Termes privilégiés ==
=== ===

 
== Anglais ==

=== New_english_term ===

Architecture par microlots

2018-03-24T13:37:17Z

Gdpelletier : Page vide créée

Développeur de plateforme

2018-03-24T12:02:04Z

Gdpelletier : Page créée avec « __NOTOC__ == Domaine == Category:VocabularyVocabulary == Définition == Platform Builder spend about half of their time building platforms and instrum... »

__NOTOC__

== Domaine ==
[[Category:Vocabulary]]Vocabulary 

 

== Définition ==
Platform Builder spend about half of their time building platforms and instrumenting code for the purpose of collecting data. Platform Builders are more likely to work in distributed systems, like Hadoop, and have “engineer” in their title, but not to have a PhD.
 
== Termes privilégiés ==
=== ===

 
== Anglais ==

=== New_english_term ===