Not sure how deep down the rabbit hole you want to go, but you might want to take a look at convolutional neural networks and their applicability to go, e.g.: http://arxiv.org/abs/1412.6564
They are used ubiquitously for image classification and object detection, but people are looking at tying them to Go as well. They have some analogs to biological image processing (gabor filters, etc.), but are not 1-to-1. The tricky thing with respect to visualization, is that while visualizations of lower-level layers in the network may make some sense, higher-level layers don't visualize as nicely. Take a look at NVIDIA's DIGITS project to see how they are visualizing. On Sat, Aug 1, 2015 at 12:46 AM, djhbrown . <djhbr...@gmail.com> wrote: > i'm looking for people with expertise in > art, > computer graphics > , > and/or AI software to help me make a video series about mental imagery in > Go, partly with a view to expressing mental images in the form of heuristic > rules that a program could use to generate and evaluate candidate moves. > > any takers? > > https://www.youtube.com/watch?v=Om_CVAevU8o > > -- > personal website <http://sites.google.com/site/djhbrown2/home> > > _______________________________________________ > Computer-go mailing list > Computer-go@computer-go.org > http://computer-go.org/mailman/listinfo/computer-go >
_______________________________________________ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
