FlowGenesis：基于表征变化自动增删神经元的自适应架构框架
FlowGenesis: Representation-Driven Self-Organization of Neural Architectures
arXiv 上这篇论文挺有意思：网络自己决定加神经元还是砍神经元，不用预先定架构，持续学习场景还能省参数，做 NAS 的可以看看。
arXiv 论文提出 FlowGenesis，用 Representation Flux 这一局部统计量追踪 minibatch 平均激活的变化，指导神经元的分裂与剪枝。高 Flux 的神经元分裂、低 Flux 且低活跃的神经元被剪除，网络容量无需预先指定或单独做架构搜索。在图像与文本分类实验中，不同初始大小的网络收敛到相近的神经元数量区间，预测性能相当。扩展到持续学习后，该机制在任务增量学习设置下减少了分配给后续任务的神经元与参数量，准确率与固定宽度的 Progressive Neural Networks 相当。
FlowGenesis: Representation-Driven Self-Organization of Neural Architectures
Deep neural networks typically rely on architectures specified before training, despite the continuous evolution of their internal representations. We introduce FlowGenesis, a framework in which neural architectures adapt through neuron birth and pruning guided by Representation Flux, a local statistic capturing changes in minibatch mean activations. Neurons exhibiting persistently high Flux undergo splitting designed to approximately preserve the network function, while neurons with low Flux and low activity are pruned. This mechanism allows network capacity to evolve without prescribing a target architecture or performing a separate architecture search. Experiments on image and text classi- fication show that networks initialized with different sizes approach similar final neuron-count ranges within each dataset while maintaining comparable predictive performance. We further extend FlowGenesis to continual learning, where representation-guided structural adaptation reduces the neurons and parameters allocated to successive tasks while maintaining accuracy comparable to fixed-width Progressive Neural Networks in a task-incremental learning setting. These findings demonstrate that neural architectures can adapt through feedback between representation learning and structural change, suggesting an alternative to prescribing network capacity before training.