Defects of Convolutional Decoder Networks in Frequency Representation

no code implementations • 17 Oct 2022 • Ling Tang, Wen Shen, Zhanpeng Zhou, Yuefeng Chen, Quanshi Zhang

In this paper, we prove the representation defects of a cascaded convolutional decoder network, considering the capacity of representing different frequency components of an input sample.

Decoder

Batch Normalization Is Blind to the First and Second Derivatives of the Loss

no code implementations • 30 May 2022 • Zhanpeng Zhou, Wen Shen, Huixin Chen, Ling Tang, Quanshi Zhang

In this paper, we prove the effects of the BN operation on the back-propagation of the first and second derivatives of the loss.

Novel structural evolution of several nanolaminate Mn+1AXn (n=1, 2, 3, etc.) ceramics under pressure from first principles

no code implementations • 22 Dec 2020 • Ben-Yang Li, Fang Chen, Heng-Na Xiong, Ling Tang, Ju-Xiang Shao, Ze-Jin Yang

We did extensive research for the typical nanolaminate Mn+1AXn (n=1, 2, 3) ceramics focusing on the structural stability, the phase transition pressure of Ti2GaN (160 GPa) is far higher than that of Zr2GaN (92 GPa), meaning the strong M dependence of the same group, whereas Zr2AlN (98 GPa) has similar value with that of Zr2GaN, meaning the weak A dependence.

Materials Science