In a paper I once read from which I now possess only a screenshot, I saw the following proof in an appendix:
Assume not $Z_1 \neq Z_2$
Expanding the definitions, we see that $Z_1 = Z_2$
This is a contradiction.
Thus, $Z_1 = Z_2$
Where $Z_1, Z_2$ were the diffusion-convolution activations of two isomorphic graphs. (These words don’t mean anything to me, I’m just reading my screenshot.)
I think I better understand the process that generated this proof.
In a paper I once read from which I now possess only a screenshot, I saw the following proof in an appendix:
Where $Z_1, Z_2$ were the diffusion-convolution activations of two isomorphic graphs. (These words don’t mean anything to me, I’m just reading my screenshot.)
I think I better understand the process that generated this proof.