Collaborative Learning
Collaborative learning focuses on improving machine learning model performance by leveraging data and computational resources from multiple sources, enhancing efficiency and privacy. Current research emphasizes developing algorithms that optimize communication and resource allocation among collaborating agents, including bilevel optimization and spectral estimators, while addressing challenges like negative transfer and data heterogeneity across diverse datasets. This approach is significant for improving model accuracy and generalization in various applications, from medical image segmentation to personalized recommendations, and is increasingly relevant in the context of data privacy regulations and resource-constrained environments.
Papers
May 5, 2022
April 27, 2022
March 21, 2022
March 19, 2022
February 21, 2022
January 31, 2022
January 29, 2022
January 23, 2022
January 14, 2022
December 7, 2021
December 2, 2021