This was an experiment and I no longer think bayes rule is the right way. GRU and LSTM and RBM neuralnets combined with some variety of computing theory (such as lambdas in occamsfuncer or neural turing machine with sparse memory, etc) are better. But this is still interesting to play with.

Version 0.1.5 decays bayesian weights toward mostly prediction of each child based on the other 2 childs, for each of 3 childs. The hard-coded brainwave is barely seen and an imagined brainwave crawls like an inchworm, stretching its green center which matches the green hard-coded brainwave. This appears to be the result of a kind of physics in the equations of how brainwaves flow and stretch. Its starting to look like an artificial intelligence’s dream, but we won’t get into that much until many people connect to eachother through it on the Internet.

A paint program where the canvas is the visual cortex of a simple kind of artificial intelligence. You paint with the mouse into its dreams and it responds by changing what you painted gradually. There will also be an API for using it with other programs as a general high-dimensional space. Each pixel’s brightness is its own dimension. Bayesian nodes have exactly 3 childs because that is all thats needed to do NAND in a fuzzy way as Bayes’ Rule which is NAND at certain extremes. NAND can be used to create any logical system. In this early version, I’m still working on edge detection and its understanding of the same shapes at different brightnesses. This will be a module of the bigger Human AI Net project and will be used for adding realtime intuitive high dimensional intelligence in audio and visual interactions with the user.

Features
Later versions will have more intelligence, logic, creativity in their responses.
Version 0.1.5 decays bayesian weights toward mostly prediction of each child based on the other 2 childs, for each of 3 childs. The hard-coded brainwave is barely seen and an imagined brainwave crawls like an inchworm, stretching its green center which matches the green hard-coded brainwave. This appears to be the result of a kind of physics in the equations of how brainwaves flow and stretch. Its starting to look like an artificial intelligence’s dream, but we won’t get into that much until many people connect to eachother through it on the Internet.
Version 0.1.4 uses red, green, and blue to show properties of each pixel node, which are leaf nodes and are connected by 3-way nodes. After 10 seconds the small effect of horizontal vibrating lines seen in version 0.1.3 happens very strongly and in all directions, affected by patterns of things you draw. See 0.1.3 to understand it in a simpler way, since 0.1.4 is more to show how the program will be fun, and 0.1.3 is to show how that part works. The subtle thing to look for in version 0.1.4 is, since red is standard deviation, green is attention, and blue is accuracy, that it starts very red, gradually becomes mostly blue, and only then starts flashing in a psychedelic way.
Version 0.1.3 demonstrates a constant speed brainwave generating an optical illusion (fooling itself, not you watching it) of the brainwave jumping around and vibrating horizontal lines after the brainwave trains it into that pattern after a minute.
Version 0.1.2 demonstrates 3 vertical and 3 horizontal bayesian NAND constraints in a grid of 9 large pixels which can be increased or decreased in brightness with the mouse. These 3-way nodes can be between any pixels or hidden variables but are vertical and horizontal in this case.
Version 0.1.1 mostly demonstrates how painting works and a constant speed brainwave that doesn’t do much.