guanchengwan/s3gcl

[ICML 2024] S3GCL: Spectral, Swift, Spatial Graph Contrastive Learning

Hi 👋

I'm SuHui, currently an undergraduate student at the School of Computer Science, Wuhan University (WHU), China.

🔎 Research

"All things are interconnected, this is the essence of nature."

I am enthusiastic about modeling the relationship of these connections, particularly focusing on geometric data. Currently, my research interests revolve around three primary areas:

a) AI for Science:

• Biotechnology (e.g., Genomics)

• Public Health (e.g., Epidemiology)

• Physics (e.g., Neural ODE, PINN)

b) TrustWorthy Graph Learning

c) Self-supervised Graph Learning

🪐Devices

s3gcl's Issues

请教问题:

你好,这篇文章实验/理论都不错.但是我有一个问题想请教一下:
在我个人看来,切比雪夫过滤器的系数为正时,是高通过滤器.(举例: $T_{1}(L)=x,T_{2}(L)=2x^2-1,T_{3}(x)=4x^3-3x$ ,都是高通过滤器).但是,你构造的两个过滤器(高通过滤器和低通过滤器)的系数都是正的,我觉得他俩都是高通过滤器.(个人理解,是理解错了吗?)
希望可以答疑解惑~