Skip to main content
Artificial Intelligence

Return to Question

added 147 characters in body; edited tags
Source Link
nbro
nbro
  • 41.4k
  • 12
  • 114
  • 205

rotational invariance Why the non-exploitation of filtersedge labels in regularcurrent graph convolutions - Neural Networks on Graphs"results in an overly homogeneous view of local graph neighborhoods"?

I am currently reading a paper called "Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs"Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs (2017, CPPR), and iI cannot understand the following sentence:

We identify that the current formulations of graph convolution do not exploit edge labels, which results in an overly homogeneous view of local graph neighborhoods, with an effect similar to enforcing rotational invariance of filters in regular convolutions on images.

What does this sentence mean?

rotational invariance of filters in regular convolutions - Neural Networks on Graphs

I am currently reading paper called "Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs" and i cannot understand the following sentence:

We identify that the current formulations of graph convolution do not exploit edge labels, which results in an overly homogeneous view of local graph neighborhoods, with an effect similar to enforcing rotational invariance of filters in regular convolutions on images.

What does this sentence mean?

Why the non-exploitation of edge labels in current graph convolutions "results in an overly homogeneous view of local graph neighborhoods"?

I am currently reading a paper called Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs (2017, CPPR), and I cannot understand the following sentence:

We identify that the current formulations of graph convolution do not exploit edge labels, which results in an overly homogeneous view of local graph neighborhoods, with an effect similar to enforcing rotational invariance of filters in regular convolutions on images.

What does this sentence mean?

Source Link
S.Rucinski
S.Rucinski
  • 13
  • 2

rotational invariance of filters in regular convolutions - Neural Networks on Graphs

I am currently reading paper called "Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs" and i cannot understand the following sentence:

We identify that the current formulations of graph convolution do not exploit edge labels, which results in an overly homogeneous view of local graph neighborhoods, with an effect similar to enforcing rotational invariance of filters in regular convolutions on images.

What does this sentence mean?