Consensus Graph
Consensus graph learning aims to integrate information from multiple data sources (views) to create a unified representation that captures shared underlying structure. Current research focuses on developing efficient algorithms, often incorporating anchor graphs or graph neural networks, to construct this consensus graph, simultaneously addressing challenges like computational complexity and data heterogeneity. This approach is proving valuable in diverse applications, including multi-view clustering, semi-supervised learning, and improving the robustness of models in areas such as image captioning and federated learning by leveraging the shared information across different views to enhance performance and generalization.
Papers
September 25, 2024
August 11, 2024
January 30, 2024
January 24, 2024
September 14, 2023
June 2, 2023
January 17, 2023
October 13, 2022
July 4, 2022
April 7, 2022
December 2, 2021