Abstract
Comprehensively consider model performance, parameter storage space, and model training time.Disintegrate the model into Shared and Private Latent Representations through Synaptic Intelligence.Achieve similar performance with comparative lifelong learning algorithms.Obtain almost the minimum training time while considering the parameter quantity.Learn the capacity of shared and private modules and sparse parameters from the dataset.
Yang, Y. , Huang, J. , & Hu, D. . (2023). Lifelong learning with shared and private latent representations learned through synaptic intelligence. Neural Networks,163, 165-177.
