- Java Deep Learning Projects
- Md. Rezaul Karim
- 124字
- 2025-04-04 16:36:26
Residual neural networks
Since there are sometimes millions of billions of hyperparameters and other practical aspects, it's really difficult to train deeper neural networks. To overcome this limitation, Kaiming He et al. (see https://arxiv.org/abs/1512.03385v1) proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously.
They also explicitly reformulated the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. This way, these residual networks are easier to optimize and can gain accuracy from considerably increased depth.
The downside is that building a network by simply stacking residual blocks inevitably limits its optimization ability. To overcome this limitation, Ke Zhang et al. also proposed using a Multilevel Residual Network (https://arxiv.org/abs/1608.02908).