Trjtdtk

Field Length Validation for Desktop Application which has maximum 1000 characters

Hang 20lb projector screen on Hardieplank

Transfer over $10k

Pigeonhole Principle Problem

Was Hulk present at this event?

Write to EXCEL from SQL DB using VBA script

Why is this a valid proof for the harmonic series?

Can a cyclic Amine form an Amide?

How did Arya manage to disguise herself?

Proof that when f'(x) < f(x), f(x) =0

Would "lab meat" be able to feed a much larger global population

Binary Numbers Magic Trick

Has any spacecraft ever had the ability to directly communicate with civilian air traffic control?

How to creep the reader out with what seems like a normal person?

How did Captain America use this power?

What word means "to make something obsolete"?

Airbnb - host wants to reduce rooms, can we get refund?

Historically, were women trained for obligatory wars? Or did they serve some other military function?

If an enemy is just below a 10-foot-high ceiling, are they in melee range of a creature on the ground?

Password expiration with Password manager

How to efficiently calculate prefix sum of frequencies of characters in a string?

CRT Oscilloscope - part of the plot is missing

Entropy as a function of temperature: is temperature well defined?

You look catfish vs You look like a catfish?

Struggling to understand GCN (Graph Convolutional Networks)

Partitioning Weighted Undirected GraphCanonical VS Graph IsomorphismConvolutional Neural Networks for Sensor DataWhat is deconvolution operation used in Fully Convolutional Neural Networks?What are graph embedding?Convolutional Neural networksConvolutional Neural Networks layer sizesUnderstand the shape of this Convolutional Neural NetworkWhy don't convolutional computer vision networks use horizontally - symmetric filters?Pretrained graph embeddings

0

1

$begingroup$

Although I've worked with CNN's for over a year, I am struggling to understand how GCN's work. I've read several papers, and I find myself out of my depth when they talk about Chebyshev polynomials or Fourier spaces.

The descriptions talk about using an adjacency matrix as input, and perhaps my primary confusion is how I can supply such a matrix to a convolutional neural network (if that is what is in fact what is done). I can't just convolve over the matrix as if it were an image because spatial similarity in the matrix (i.e. rows/cols that are near to each other) doesn't signify actual closeness between nodes in the graph.

Can anyone clarify this?

neural-network convolution graphs

share|improve this question

asked Apr 9 at 0:18

JellicleCatJellicleCat

1012

$endgroup$

add a comment | 

0

1

$begingroup$

Although I've worked with CNN's for over a year, I am struggling to understand how GCN's work. I've read several papers, and I find myself out of my depth when they talk about Chebyshev polynomials or Fourier spaces.

The descriptions talk about using an adjacency matrix as input, and perhaps my primary confusion is how I can supply such a matrix to a convolutional neural network (if that is what is in fact what is done). I can't just convolve over the matrix as if it were an image because spatial similarity in the matrix (i.e. rows/cols that are near to each other) doesn't signify actual closeness between nodes in the graph.

Can anyone clarify this?

neural-network convolution graphs

share|improve this question

asked Apr 9 at 0:18

JellicleCatJellicleCat

1012

$endgroup$

add a comment | 

$begingroup$

Although I've worked with CNN's for over a year, I am struggling to understand how GCN's work. I've read several papers, and I find myself out of my depth when they talk about Chebyshev polynomials or Fourier spaces.

The descriptions talk about using an adjacency matrix as input, and perhaps my primary confusion is how I can supply such a matrix to a convolutional neural network (if that is what is in fact what is done). I can't just convolve over the matrix as if it were an image because spatial similarity in the matrix (i.e. rows/cols that are near to each other) doesn't signify actual closeness between nodes in the graph.

Can anyone clarify this?

neural-network convolution graphs

share|improve this question

asked Apr 9 at 0:18

JellicleCatJellicleCat

1012

$endgroup$

share|improve this question

asked Apr 9 at 0:18

JellicleCatJellicleCat

1012

share|improve this question

asked Apr 9 at 0:18

JellicleCatJellicleCat

1012

asked Apr 9 at 0:18

JellicleCatJellicleCat

1012

asked Apr 9 at 0:18

JellicleCatJellicleCat

1012