郑之杰的个人博客
欢迎光临
On the Parameterization and Initialization of Diagonal State Space Models
对角状态空间模型的参数化和初始化.
Simplified State Space Layers for Sequence Modeling
简化序列建模的状态空间层.
Diagonal State Spaces are as Effective as Structured State Spaces
对角状态空间和结构化状态空间一样有效.
Resurrecting Recurrent Neural Networks for Long Sequences
复活长序列的循环神经网络.
Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layers
通过线性状态空间层结合循环、卷积核连续时间模型.
Legendre Memory Units: Continuous-Time Representation in Recurrent Neural Networks
勒让德记忆单元:循环神经网络中的连续时间表示.