Sélection de la langue

Gouvernement du Canada / Government of Canada

Recherche

Aggregated learning: a vector-quantization approach to learning neural network classifiers

Par Conseil national de recherches du Canada

AuteurRechercher : ; Rechercher : 1; Rechercher : ; Rechercher : ; Rechercher :
Affiliation
  1. Conseil national de recherches du Canada. Technologies numériques
FormatTexte, Article
Conférence34th AAAI Conference on Artificial Intelligence, AAAI 2020, February 7-12, 2020, New York, USA
Sujetcharacter recognition; classification (of information); electric distortion; equivalence classes; image coding; image recognition; signal distortion; text processing; variational techniques; vector quantization; neural networks
Résumé
We consider the problem of learning a neural network classifier. Under the information bottleneck (IB) principle, we associate with this classification problem a representation learning problem, which we call “IB learning”. We show that IB learning is, in fact, equivalent to a special class of the quantization problem. The classical results in rate-distortion theory then suggest that IB learning can benefit from a “vector quantization” approach, namely, simultaneously learning the representations of multiple input objects. Such an approach assisted with some variational techniques, result in a novel learning framework, “Aggregated Learning”, for classification with neural network models. In this framework, several objects are jointly classified by a single neural network. The effectiveness of this framework is verified through extensive experiments on standard image recognition and text classification tasks.
Date de publication
Maison d’éditionAAAI press
Dans
  • AAAI 2020: 34th AAAI Conference on Artificial Intelligence (7 février 2020) : 5810–5817.
Langueanglais
Publications évaluées par des pairsOui
Exporter la noticeExporter en format RIS
Signaler une correctionSignaler une correction (s'ouvre dans un nouvel onglet)
Identificateur de l’enregistrementfea97c89-ad6b-4696-8a2f-147e402bcf38
Enregistrement créé2022-07-22
Enregistrement modifié2022-07-22
Date de modification :