Always private
DuckDuckGo never tracks your searches.
Learn More
You can hide this reminder in Search Settings
All regions
Argentina
Australia
Austria
Belgium (fr)
Belgium (nl)
Brazil
Bulgaria
Canada (en)
Canada (fr)
Catalonia
Chile
China
Colombia
Croatia
Czech Republic
Denmark
Estonia
Finland
France
Germany
Greece
Hong Kong
Hungary
Iceland
India (en)
Indonesia (en)
Ireland
Israel (en)
Italy
Japan
Korea
Latvia
Lithuania
Malaysia (en)
Mexico
Netherlands
New Zealand
Norway
Pakistan (en)
Peru
Philippines (en)
Poland
Portugal
Romania
Russia
Saudi Arabia
Singapore
Slovakia
Slovenia
South Africa
Spain (ca)
Spain (es)
Sweden
Switzerland (de)
Switzerland (fr)
Taiwan
Thailand (en)
Turkey
Ukraine
United Kingdom
US (English)
US (Spanish)
Vietnam (en)
Safe search: moderate
Strict
Moderate
Off
Any time
Any time
Past day
Past week
Past month
Past year
  1. ieeexplore.ieee.org

    Jul 2, 2024Recent deep learning models can efficiently combine inputs from different modalities (e.g., images and text) and learn to align their latent representations or to translate signals from one domain to another (as in image captioning or text-to-image generation). However, current approaches mainly rely on brute-force supervised training over large multimodal datasets. In contrast, humans (and ...
  2. Here, we explore the capabilities of a multimodal system taking inspiration from the cognitive science theory of the Global Workspace (GW) [14, 15].GW Theory explains how different modalities in the human brain are integrated into a common shared representation, subsequently redistributed or broadcast among the specialized unimodal modules (see section II-B).
  3. combining a global workspace with semi-supervision, in a bimodal setting where only few paired examples are available. II. RELATED WORK As explained above, multimodal representation learning for neural networks is a vast and fast-growing research area, whose exhaustive coverage would require an extensive review well beyond the scope of the ...
  4. semanticscholar.org

    Jun 27, 2023It is shown that an architecture inspired by the cognitive notion of a "global workspace" (GW) can be trained to align and translate between two modalities with very little need for matched data (from four to seven times less than a fully supervised approach). Recent deep learning models can efficiently combine inputs from different modalities (e.g., images and text) and learn to align their ...
  5. researchgate.net

    Jun 27, 2023Download Citation | Semi-supervised Multimodal Representation Learning through a Global Workspace | Recent deep learning models can efficiently combine inputs from different modalities (e.g ...
  6. semanticscholar.org

    Jun 27, 2023The goal of this tensor is to capture the fixed background such that only the changing image elements (i.e., the table and its immediate suroundings) are passed to the VAE. B) Evolution of X0 through successive training epochs t. - "Semi-supervised Multimodal Representation Learning through a Global Workspace"
  7. Jun 27, 2023Recent deep learning models can efficiently combine inputs from different modalities (e.g., images and text) and learn to align their latent representations, or to translate signals from one domain to another (as in image captioning, or text-to-image generation). However, current approaches mainly rely on brute-force supervised training over large multimodal datasets. In contrast, humans (and ...
  8. The global workspace representation can be used advantageously for downstream classification tasks and for robust transfer learning. Ablation studies reveal that both the shared workspace and the self-supervised cycle-consistency training are critical to the systems performance.
  9. pubmed.ncbi.nlm.nih.gov

    In contrast, humans (and other animals) can learn useful multimodal representations from only sparse experience with matched cross-modal data. Here, we evaluate the capabilities of a neural network architecture inspired by the cognitive notion of a "global workspace" (GW): a shared representation for two (or more) input modalities.
  10. paperswithcode.com

    Jun 27, 2023In contrast, humans (and other animals) can learn useful multimodal representations from only sparse experience with matched cross-modal data. Here we evaluate the capabilities of a neural network architecture inspired by the cognitive notion of a "Global Workspace": a shared representation for two (or more) input modalities.

    Can’t find what you’re looking for?

    Help us improve DuckDuckGo searches with your feedback

Custom date rangeX