近期关于感知器MLP的最新研究

近期关于感知器MLP的最新研究

1. MLP-Mixer: An all-MLP Architecture for Vision

Google Research, Brain Team团队的工作
文章链接: https://arxiv.org/pdf/2105.01601.pdf
代码链接:https://github.com/google-research/vision_transformer/tree/linen

2. ResMLP: Feedforward networks for image classification with data-efficient training

Facebook AI团队的工作
文章链接:https://arxiv.org/abs/2105.03404
代码链接:https://github.com/lucidrains/res-mlp-pytorch

3. Do You Even Need Attention? A Stack of Feed-Forward Layers Does Surprisingly Well on ImageNet

牛津大学团队的工作
文章链接:https://arxiv.org/abs/2105.02723
代码链接:https://github.com/lukemelas/do-you-even-need-attention

4. RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition

清华大学&旷视科技团队联合工作
文章链接:https://arxiv.org/abs/2105.01883
代码链接:https://github.com/DingXiaoH/RepMLP

上一篇:tf.contrib.layers.batch_norm 报错


下一篇:TCP/IP capture/analysis tools in Unix/Linux