Transferring Core Knowledge via Learngenes
This offers a more efficient transfer learning method for neural network practitioners, though it is incremental as it builds on existing concepts like learngenes.
The paper tackles the inefficiency of transferring all knowledge in pre-trained models by proposing Genetic Transfer Learning (GTL), which extracts and transfers only essential 'learngenes' from networks like VGG11 and ResNet12, resulting in 12% and 16% accuracy improvements on CIFAR-FS and miniImageNet with only 20% of parameters.
The pre-training paradigm fine-tunes the models trained on large-scale datasets to downstream tasks with enhanced performance. It transfers all knowledge to downstream tasks without discriminating which part is necessary or unnecessary, which may lead to negative transfer. In comparison, knowledge transfer in nature is much more efficient. When passing genetic information to descendants, ancestors encode only the essential knowledge into genes, which act as the medium. Inspired by that, we adopt a recent concept called ``learngene'' and refine its structures by mimicking the structures of natural genes. We propose the Genetic Transfer Learning (GTL) -- a framework to copy the evolutionary process of organisms into neural networks. GTL trains a population of networks, selects superior learngenes by tournaments, performs learngene mutations, and passes the learngenes to next generations. Finally, we successfully extract the learngenes of VGG11 and ResNet12. We show that the learngenes bring the descendant networks instincts and strong learning ability: with 20% parameters, the learngenes bring 12% and 16% improvements of accuracy on CIFAR-FS and miniImageNet. Besides, the learngenes have the scalability and adaptability on the downstream structure of networks and datasets. Overall, we offer a novel insight that transferring core knowledge via learngenes may be sufficient and efficient for neural networks.