Live data from Hacker News

Show HN: Graph Convolutional Networks – Intro to neural networks on graphs

tkipf.github.io

1–10 of 13 posts

Re: Show HN: Graph Convolutional Networks – Intro to neural networks on graphs

#3

Does this classify whole graphs, or nodes into communities/classes?

Both graph-level and node-level classification are possible. Graph-level classification requires some from of pooling operation (simplest case: mean-pooling over all nodes, but there are more elaborate things one can do)

Re: Show HN: Graph Convolutional Networks – Intro to neural networks on graphs

#5
post #4

This looks really interesting, but I'm struggling to think of a practical application. Any ideas?

I know little, but how about classification of neuro–developmental disorders from network graphs derived from resting state fMRI?

Re: Show HN: Graph Convolutional Networks – Intro to neural networks on graphs

#7
post #6

How does this differ from graphical models like factor graphs ?

In a graphical model, you'd explicitly model the probabilistic assumptions that you make with respect to the data. In this neural network-based approach the goal can be thought of more like learning a function that maps from some input to some desired output. But indeed the form of the propagation rule resembles mean field inference in graphical models.

Re: Show HN: Graph Convolutional Networks – Intro to neural networks on graphs

#8
post #4

This looks really interesting, but I'm struggling to think of a practical application. Any ideas?

I know little, but how about classification of neuro–developmental disorders from network graphs derived from resting state fMRI?

Stupid question: can bayesian networks (graphs) be a useful input?

Re: Show HN: Graph Convolutional Networks – Intro to neural networks on graphs

#9

Earlier quoted context omitted.

I know little, but how about classification of neuro–developmental disorders from network graphs derived from resting state fMRI?

Stupid question: can bayesian networks (graphs) be a useful input?

Currently the framework only supports undirected graphs, so directed graphical models wouldn't be supported as input. I can't really judge how useful it would be to take a Bayesian net as input, sounds a bit hacky to me. But in principle you could train a neural net on any kind of graph, someone recently suggested to take the connection graph of another neural net as input and try to learn some function on that. But again, it's really hard to judge in advance how useful such approaches would be ;)

Re: Show HN: Graph Convolutional Networks – Intro to neural networks on graphs

#10
This is really powerful. As a use case I'm considering the graph of streets in a city. A task would be to make predictions on the next nodes visited by a vehicle given the history of the recent nodes. I'm not sure how you deal with mini batches. Since you need the Adj matrix of the whole graph?
Post reply on HN