郑之杰的个人博客
欢迎光临
Dynamic Task Prioritization for Multitask Learning
多任务学习中的动态任务优先级.
Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
使用稀疏门控的混合专家系统构建超大规模神经网络.
Region-based Non-local Operation for Video Classification
为视频分类设计的基于区域的非局部网络.
Exploring Self-attention for Image Recognition
探索图像识别的自注意力机制.
Image Super-Resolution with Non-Local Sparse Attention
通过非局部稀疏注意力实现图像超分辨率.
Polarized Self-Attention: Towards High-quality Pixel-wise Regression
极化自注意力: 面向高质量像素级回归.