Talk:Capsule neural network

I've finished with this piece now. It is ready for submission and/or publishing. Feedback encouraged. Lfstevens (talk) 02:02, 29 December 2017 (UTC)[reply]

In my version I had carefully split out intuition and motivation. As it is now the article seems like the whole idea emerged from nowhere. Those that has follwed the field for some time knows that this idea has a history long before the latest papers.

Intuition v motivation? I have no opinion about that subject, but we have to have sources to talk about it. I would love to expand the history bit. I just need sources.

Now the article is about visual representation of objects, which is kind of weird. First of all, spatial relationship between parts of an object is not the object itself. This is about information represented in a vector space, and how transformations on the information can reconnect otherwise disparate points in that space. That can be used for reconnecting visual parts of an object, but that is only one of several use cases.

I find the non-visual aspect intriguing (once you have vectors, they need not be limited to geometric pose issues). But we need sources. What encourages me about capsnets is that now objects have properties. OK, only scalar properties and only hierarchical relationships, but that can evolve.

In my version I wrote about how it compares to cortical minicolumns, now it is about a special case where capsule networks are used for interpreting visual scenes. My version was about how capsule nets can be interpreted in the general case, now it is about a special solution for a special problem.

I don't think I deleted anything about minicolumns and am happy to restore anything that got lost. I agree that capsnets could be useful for data other than images. But we need sources that talk about that.

Nearly everything about the inspiration from cortical minicolumns are gone, which is very bad, as it removes the foundation for why capsule networks have some serious flaws. The flaws is how the dynamic routing is done, actually the error is the routing. Inspecting the history it is clear that you simply dismissed that as speculation. Well, to put it bluntly, capsule nets are not how the brain does it, but it is a small step in the right direction.

I'm not disagreeing with your conclusions, but they have to be supported by reliable sources, rather than your insights. Is your objection that capsnets don't follow the brain, or that they are a bad idea in general?

No, I don't think this draft should be posted as it is now. Jeblad (talk) 00:13, 30 December 2017 (UTC)[reply]

Happy to keep working on this. Is there anything in the piece now that is incorrect? Lfstevens (talk) 07:30, 30 December 2017 (UTC)[reply]