<html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"><head><meta http-equiv=Content-Type content="text/html; charset=utf-8"><meta name=Generator content="Microsoft Word 15 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:#954F72;
text-decoration:underline;}
.MsoChpDefault
{mso-style-type:export-only;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style></head><body lang=EN-US link=blue vlink="#954F72"><div class=WordSection1><p class=MsoNormal>We are speaking of the difference between a HyperMovie and a HyperShape, at a fundamental level. One is appearance, the other is geometry. What is the appropriate type of neural network for handling geometry?</p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Yes, I realize it’s pixels when you look at it. However, a shape is not always rectangular or flat. This is applying neural networks to Non-Euclidean data (or so they say).</p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Seems like a very complex problem how to feed various geometries to a neural network. I would tend toward rectangularizing it.</p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>So similar to how a CNN traverses through an image finding features, a GCN or HCN traverses through a graph or hypergraph finding features. That’s the main difference. The similarity is that they’re all convolutional.</p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>I agree that images can be graphs and graphs can be images.</p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Hmm.</p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>John</p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Sent from <a href="https://go.microsoft.com/fwlink/?LinkId=550986">Mail</a> for Windows 10</p><p class=MsoNormal><o:p> </o:p></p><div style='mso-element:para-border-div;border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in'><p class=MsoNormal style='border:none;padding:0in'><b>From: </b><a href="mailto:yottzumm@gmail.com">John Carlson</a><br><b>Sent: </b>Wednesday, May 22, 2019 11:02 AM<br><b>To: </b><a href="mailto:joedwil@earthlink.net">Joseph D Williams</a>; <a href="mailto:x3d-public@web3d.org">X3D Graphics public mailing list</a><br><b>Subject: </b>RE: [x3d-public] Geometric/Geographic Deep Learning</p></div><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>So instead of dealing with output from neural networks being voxels, maybe, just maybe, we can have graphs and meshes? I’m not entirely clear on the distinction between the data and the network. I guess a GCN can take graphs as input?<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><a href="https://arxiv.org/pdf/1903.10384.pdf">https://arxiv.org/pdf/1903.10384.pdf</a><o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>John<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Sent from <a href="https://go.microsoft.com/fwlink/?LinkId=550986">Mail</a> for Windows 10<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><div style='border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in'><p class=MsoNormal><b>From: </b><a href="mailto:yottzumm@gmail.com">John Carlson</a><br><b>Sent: </b>Wednesday, May 22, 2019 10:44 AM<br><b>To: </b><a href="mailto:joedwil@earthlink.net">Joseph D Williams</a>; <a href="mailto:x3d-public@web3d.org">X3D Graphics public mailing list</a><br><b>Subject: </b>RE: [x3d-public] Geometric/Geographic Deep Learning<o:p></o:p></p></div><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>No, you don’t get it. It’s not even a picture/image/frame. It’s a graph/mesh. Not a CNN. A GCN or HCN. No one said anything about moving or frames except you.<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>In other words, we’ve gotten past pixels in our thinking.<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>John<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Sent from <a href="https://go.microsoft.com/fwlink/?LinkId=550986">Mail</a> for Windows 10<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><div style='border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in'><p class=MsoNormal><b>From: </b><a href="mailto:joedwil@earthlink.net">Joseph D Williams</a><br><b>Sent: </b>Wednesday, May 22, 2019 10:36 AM<br><b>To: </b><a href="mailto:yottzumm@gmail.com">John Carlson</a>; <a href="mailto:x3d-public@web3d.org">X3D Graphics public mailing list</a><br><b>Subject: </b>RE: [x3d-public] Geometric/Geographic Deep Learning<o:p></o:p></p></div><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><a href="https://www.youtube.com/watch?v=aircAruvnKk">https://www.youtube.com/watch?v=aircAruvnKk</a><o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>When things are moving, we can start to think of frames. If no movement, only one frame needed. <o:p></o:p></p><p class=MsoNormal>Well, we start with the idea of a network, then thinking about how to invent a computing structure to compute the stuff.<o:p></o:p></p><p class=MsoNormal>The hardware and the training seems to be very important. That phrase of continuing integration holds the idea of a dynamically changing output result. <o:p></o:p></p><p class=MsoNormal>Joe<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p><div style='border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in'><p class=MsoNormal><b>From: </b><a href="mailto:yottzumm@gmail.com">John Carlson</a><br><b>Sent: </b>Sunday, May 19, 2019 12:18 PM<br><b>To: </b><a href="mailto:joedwil@earthlink.net">Joseph D Williams</a>; <a href="mailto:x3d-public@web3d.org">X3D Graphics public mailing list</a><br><b>Subject: </b>RE: [x3d-public] Geometric/Geographic Deep Learning<o:p></o:p></p></div><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Uh, I just wanted to do geometric and geographic deep learning?<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>“Frame”? <a href="https://www.youtube.com/watch?v=D3fnGG7cdjY">https://www.youtube.com/watch?v=D3fnGG7cdjY</a><o:p></o:p></p><p class=MsoNormal>John<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Sent from <a href="https://go.microsoft.com/fwlink/?LinkId=550986">Mail</a> for Windows 10<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><div style='border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in'><p class=MsoNormal><b>From: </b><a href="mailto:joedwil@earthlink.net">Joseph D Williams</a><br><b>Sent: </b>Sunday, May 19, 2019 11:58 AM<br><b>To: </b><a href="mailto:yottzumm@gmail.com">John Carlson</a>; <a href="mailto:x3d-public@web3d.org">X3D Graphics public mailing list</a><br><b>Subject: </b>RE: [x3d-public] Geometric/Geographic Deep Learning<o:p></o:p></p></div><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Anything you wish to discuss involving anticipation, simulation, recognition, labeling, intentionality, inclusion, exclusion, semantic and physical relationships, what the computer wants to see, deep learning, and continuous integration, then watch some of this.<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><a href="https://www.youtube.com/watch?v=-b041NXGPZ8">https://www.youtube.com/watch?v=-b041NXGPZ8</a><o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>to convolve and deconvolve is basic. How many frames you want? How many neurons you got? <o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Thanks, <o:p></o:p></p><p class=MsoNormal>Joe<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p><div style='border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in'><p class=MsoNormal><b>From: </b><a href="mailto:yottzumm@gmail.com">John Carlson</a><br><b>Sent: </b>Sunday, May 19, 2019 8:34 AM<br><b>To: </b><a href="mailto:x3d-public@web3d.org">X3D Graphics public mailing list</a><br><b>Subject: </b>[x3d-public] Geometric/Geographic Deep Learning<o:p></o:p></p></div><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Finally, something that interests me about deep learning! Is anyone working on geometric or geographic deep learning? It appears like these subfields of deep learning have emerged, based on Graph Convolution Networks (GCNs), and perhaps HyperGCNs.<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Thanks,<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>John<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal><o:p> </o:p></p></div></body></html>