Pondération attentionnelle - Historique des versions

Patrickdrouin le 17 septembre 2024 à 19:15

2024-09-17T19:15:13Z

← Version précédente		Version du 17 septembre 2024 à 15:15
Ligne 27 :		Ligne 27 :

	[[Catégorie:GRAND LEXIQUE FRANÇAIS]]		[[Catégorie:GRAND LEXIQUE FRANÇAIS]]
			[[Catégorie:101]]

Patrickdrouin le 17 septembre 2024 à 19:14

2024-09-17T19:14:43Z

← Version précédente		Version du 17 septembre 2024 à 15:14
Ligne 17 :		Ligne 17 :
	'''Attention Weights'''		'''Attention Weights'''


	<!-- Attention-weighting is a technique by which the model learns which part of the incoming sequence needs to be focused on. Think of it as the ‘Eye of Sauron’ scanning everything at all times and throwing light on the parts that are relevant.

	In natural language processing, this usually means assigning different levels of importance to different words in a sentence. It assigns importance to each word by calculating "soft" weights for the word's numerical representation, known as its embedding, within a specific section of the sentence called the context window to determine its importance. The calculation of these weights can occur simultaneously in models called transformers, or one by one in models known as recurrent neural networks. Unlike "hard" weights, which are predetermined and fixed during training, "soft" weights can -->

	== Source ==		== Source ==

Pitpitt le 11 septembre 2024 à 23:48

2024-09-11T23:48:24Z

← Version précédente		Version du 11 septembre 2024 à 19:48
Ligne 30 :		Ligne 30 :


	[[Catégorie:~~Publication~~]]		[[Catégorie:GRAND LEXIQUE FRANÇAIS]]

Patrickdrouin : Patrickdrouin a déplacé la page Attention-Weighting vers Pondération attentionnelle

2024-09-10T19:07:45Z

Patrickdrouin a déplacé la page Attention-Weighting vers Pondération attentionnelle

← Version précédente	Version du 10 septembre 2024 à 15:07
(Aucune différence)

Patrickdrouin le 10 septembre 2024 à 19:03

2024-09-10T19:03:31Z

← Version précédente		Version du 10 septembre 2024 à 15:03
Ligne 1 :		Ligne 1 :
	== Définition ==		== Définition ==
	[[Mécanisme d'attention]] qui attribue un [[poids]] aux différents éléments qui constitue une [[Séquence de mots\|séquence]].		[[Mécanisme d'attention]] qui attribue un [[poids]] aux différents éléments qui constitue une [[Séquence de mots\|séquence]] en fonction de leur importance.

	== Compléments ==		== Compléments ==
	Par exemple, en [[traitement automatique de la langue]], le [[mécanisme d'attention]] attribuera des [[poids]] différents aux mots en fonction de leur ~~importance~~.		Par exemple, en [[traitement automatique de la langue]], le [[mécanisme d'attention]] attribuera des [[poids]] différents aux mots en fonction des liens entre les mots et de leur différent rôle linguistique.

	Nous proposons d'adopter le terme ''pondération attentionnelle'' qui est déjà en utilisation dans le domaine de la psychologie pour désigner un concept similaire.		Nous proposons d'adopter le terme ''pondération attentionnelle'' qui est déjà en utilisation dans le domaine de la psychologie pour désigner un concept similaire.

Patrickdrouin le 10 septembre 2024 à 19:01

2024-09-10T19:01:16Z

← Version précédente		Version du 10 septembre 2024 à 15:01
Ligne 1 :		Ligne 1 :
	== Définition ==		== Définition ==
	~~Technique qui permet au modèle~~ d'~~apprendre si~~ une ~~certaine partie~~ de la séquence ~~d'entrée nécessite une attention particulière~~.		[[Mécanisme d'attention]] qui attribue un [[poids]] aux différents éléments qui constitue une [[Séquence de mots\|séquence]].

	~~Voir aussi '''~~[[~~Mécanisme~~ d'attention]]~~'''~~		== Compléments ==
			Par exemple, en [[traitement automatique de la langue]], le [[mécanisme d'attention]] attribuera des [[poids]] différents aux mots en fonction de leur importance.

	~~== Compléments ==~~
	Nous proposons d'adopter le terme ''pondération attentionnelle'' qui est déjà en utilisation dans le domaine de la psychologie pour désigner un concept similaire.		Nous proposons d'adopter le terme ''pondération attentionnelle'' qui est déjà en utilisation dans le domaine de la psychologie pour désigner un concept similaire.

Ligne 16 :		Ligne 16 :

	'''Attention Weights'''		'''Attention Weights'''


	<!-- Attention-weighting is a technique by which the model learns which part of the incoming sequence needs to be focused on. Think of it as the ‘Eye of Sauron’ scanning everything at all times and throwing light on the parts that are relevant.		<!-- Attention-weighting is a technique by which the model learns which part of the incoming sequence needs to be focused on. Think of it as the ‘Eye of Sauron’ scanning everything at all times and throwing light on the parts that are relevant.
Ligne 23 :		Ligne 24 :
	== Source ==		== Source ==
	[https://doi.org/10.1146/annurev.psych.49.1.585 Source : Goldstone, R. L. (1998). ''Perceptual Learning. Annual Review of Psychology'']		[https://doi.org/10.1146/annurev.psych.49.1.585 Source : Goldstone, R. L. (1998). ''Perceptual Learning. Annual Review of Psychology'']

	[https://arxiv.org/pdf/2211.07714 Source : arxiv]		[https://arxiv.org/pdf/2211.07714 Source : arxiv]

	[https://en.wikipedia.org/wiki/Attention_(machine_learning) Source: wiki]		[https://en.wikipedia.org/wiki/Attention_(machine_learning) Source: wiki]


	[[Catégorie:Publication]]		[[Catégorie:Publication]]

Patrickdrouin le 10 septembre 2024 à 18:53

2024-09-10T18:53:16Z

← Version précédente		Version du 10 septembre 2024 à 14:53
Ligne 9 :		Ligne 9 :
	== Français ==		== Français ==
	''' pondération attentionnelle '''		''' pondération attentionnelle '''

	''' pondération de l'attention '''		''' pondération de l'attention '''

	== Anglais ==		== Anglais ==
	''' Attention-Weighting'''		''' Attention-Weighting'''

	'''Attention Weights'''		'''Attention Weights'''

Ligne 25 :		Ligne 27 :


	[[Catégorie:~~vocabulary~~]]		[[Catégorie:Publication]]

Patrickdrouin le 10 septembre 2024 à 18:52

2024-09-10T18:52:34Z

← Version précédente		Version du 10 septembre 2024 à 14:52
Ligne 4 :		Ligne 4 :
	Voir aussi '''[[Mécanisme d'attention]]'''		Voir aussi '''[[Mécanisme d'attention]]'''

			== Compléments ==
			Nous proposons d'adopter le terme ''pondération attentionnelle'' qui est déjà en utilisation dans le domaine de la psychologie pour désigner un concept similaire.

	== Français ==		== Français ==
	''' ~~XXXXXXXXX~~ '''		''' pondération attentionnelle '''
			''' pondération de l'attention '''

	== Anglais ==		== Anglais ==
Ligne 12 :		Ligne 15 :
	'''Attention Weights'''		'''Attention Weights'''

	Attention-weighting is a technique by which the model learns which part of the incoming sequence needs to be focused on. Think of it as the ‘Eye of Sauron’ scanning everything at all times and throwing light on the parts that are relevant.		<!-- Attention-weighting is a technique by which the model learns which part of the incoming sequence needs to be focused on. Think of it as the ‘Eye of Sauron’ scanning everything at all times and throwing light on the parts that are relevant.

	In natural language processing, this usually means assigning different levels of importance to different words in a sentence. It assigns importance to each word by calculating "soft" weights for the word's numerical representation, known as its embedding, within a specific section of the sentence called the context window to determine its importance. The calculation of these weights can occur simultaneously in models called transformers, or one by one in models known as recurrent neural networks. Unlike "hard" weights, which are predetermined and fixed during training, "soft" weights can		In natural language processing, this usually means assigning different levels of importance to different words in a sentence. It assigns importance to each word by calculating "soft" weights for the word's numerical representation, known as its embedding, within a specific section of the sentence called the context window to determine its importance. The calculation of these weights can occur simultaneously in models called transformers, or one by one in models known as recurrent neural networks. Unlike "hard" weights, which are predetermined and fixed during training, "soft" weights can -->

	== Source ==		== Source ==
			[https://doi.org/10.1146/annurev.psych.49.1.585 Source : Goldstone, R. L. (1998). ''Perceptual Learning. Annual Review of Psychology'']
	[https://arxiv.org/pdf/2211.07714 Source : arxiv]		[https://arxiv.org/pdf/2211.07714 Source : arxiv]
	[https://en.wikipedia.org/wiki/Attention_(machine_learning) Source: wiki]		[https://en.wikipedia.org/wiki/Attention_(machine_learning) Source: wiki]

Arianne le 13 août 2024 à 21:33

2024-08-13T21:33:23Z

← Version précédente		Version du 13 août 2024 à 17:33
Ligne 1 :		Ligne 1 :
	== Définition ==		== Définition ==
	~~XXXXXXXXX~~		Technique qui permet au modèle d'apprendre si une certaine partie de la séquence d'entrée nécessite une attention particulière.

			Voir aussi '''[[Mécanisme d'attention]]'''


	== Français ==		== Français ==
Ligne 7 :		Ligne 10 :
	== Anglais ==		== Anglais ==
	''' Attention-Weighting'''		''' Attention-Weighting'''
			'''Attention Weights'''

	Attention-weighting is a technique by which the model learns which part of the incoming sequence needs to be focused on. Think of it as the ‘Eye of Sauron’ scanning everything at all times and throwing light on the parts that are relevant.		Attention-weighting is a technique by which the model learns which part of the incoming sequence needs to be focused on. Think of it as the ‘Eye of Sauron’ scanning everything at all times and throwing light on the parts that are relevant.

			In natural language processing, this usually means assigning different levels of importance to different words in a sentence. It assigns importance to each word by calculating "soft" weights for the word's numerical representation, known as its embedding, within a specific section of the sentence called the context window to determine its importance. The calculation of these weights can occur simultaneously in models called transformers, or one by one in models known as recurrent neural networks. Unlike "hard" weights, which are predetermined and fixed during training, "soft" weights can

	== Source ==		== Source ==

	[https://arxiv.org/pdf/2211.07714 Source : arxiv]		[https://arxiv.org/pdf/2211.07714 Source : arxiv]
			[https://en.wikipedia.org/wiki/Attention_(machine_learning) Source: wiki]


	[[Catégorie:vocabulary]]		[[Catégorie:vocabulary]]

Pitpitt : Page créée avec « == Définition == XXXXXXXXX == Français == ''' XXXXXXXXX ''' == Anglais == ''' Attention-Weighting''' Attention-weighting is a technique by which the model learns which part of the incoming sequence needs to be focused on. Think of it as the ‘Eye of Sauron’ scanning everything at all times and throwing light on the parts that are relevant. == Source == [https://arxiv.org/pdf/2211.07714 Source : arxiv] Catégorie:vocabulary »

2024-05-06T13:35:37Z

Page créée avec « == Définition == XXXXXXXXX == Français == ''' XXXXXXXXX ''' == Anglais == ''' Attention-Weighting''' Attention-weighting is a technique by which the model learns which part of the incoming sequence needs to be focused on. Think of it as the ‘Eye of Sauron’ scanning everything at all times and throwing light on the parts that are relevant. == Source == [https://arxiv.org/pdf/2211.07714 Source : arxiv] Catégorie:vocabulary »

Nouvelle page

== Définition ==
XXXXXXXXX

== Français ==
''' XXXXXXXXX '''

== Anglais ==
''' Attention-Weighting'''

Attention-weighting is a technique by which the model learns which part of the incoming sequence needs to be focused on. Think of it as the ‘Eye of Sauron’ scanning everything at all times and throwing light on the parts that are relevant.

== Source ==

[https://arxiv.org/pdf/2211.07714 Source : arxiv]

[[Catégorie:vocabulary]]