#permutation-invariance

Timed test

2 questions · 2 min · score and explanations at the end

Q 1 / 2Score 0 / 0

Shuffle questions

Why is "flatten the $n \times n$ adjacency matrix and feed it to an MLP" a poor way to classify graphs?

Renumbering the nodes changes the input and can change the output1
An MLP cannot be trained on inputs made only of zeros and ones2
Flattening the matrix throws away the direction of directed edges3
The adjacency matrix carries no information about how the nodes are connected4

Keys: 1–9 select · Enter check / next