ArrayAlign | ArrayAlignHierarchy | ArrayAlign_ConstantRange | ArrayAlign_Equal | ArrayAlign_Multiply | ArrayAlign_PermutationCycle | ArrayAlign_Power | AudivolvNetwork | AudivolvSummary | Brainwaves | CodeSimian | CodeTree | Func | FuncPurpose_MeasureOtherFunc | FuncPurpose_SoundColorMouse | FuncPurpose_Unknown | HardCoded_CodeTree_BreedFunction | HeapQueue | HighStandards | Hypercube | Javassist | LongTermGoalsOfAudivolvDesign | MindReadingHelmet | MovementScore | NeatVsScruffy | NoSpying | Node | NormedCodeTree | PermutationCompilerAndInterpreter | Plugin | ShortTermGoalsOfAudivolvDesign | SoundCardPart | Template | TuringTests_per_second | flo | int | ob

Audivolv - LongTermGoalsOfAudivolvDesign


This is long-term. Also read ShortTermGoalsOfAudivolvDesign.

Derivatives on Hypercube vector fields.

Evolve code to predict other evolved code (all evolved code is Func).

Most musicians play instruments, but instruments should also be self-aware and play the musicians,
to form a single intelligent internet-connected system made of software and Humans using mouse and speakers,
which would use unconscious Human intelligence to evolve better evolution and music algorithms,
and create one global superintelligent brain whose brainwaves are musical,
while allowing Humans to disconnect from this brain by taking their hand off the mouse,
but being fun and useful enough that they will not want to.
It should create a different musical language specialized in communicating to each Human
that helps information on the internet and other Human's brains to flow quickly
to and from that Human's unconscious mind.
It should be a weak borg-like network without the need for brain implants.

Music and math are the universal language
and are the best way for Human and Computer to communicate.

Music is a high bandwidth connection directly to Human unconscious minds.
Moving the mouse is not as good at that, but it will work if the
Humans are motivated enough by the music to move the mouse.
The interface has to be analog, which mouse and sound are.
Keyboard will not work. It has to be simple, and speakers+mouse
is the simplest possible interface with current hardware.


I'm trying to simultaneously and in a unified way be compatible with
the most advanced forms of fuzzy AI, which are bayesian and neural networks,
but the specific types of networks should be evolved. If a node has C childs,
a neural node has C+1 floating points, and a bayesian node has 2^(C+1)
floating points. An audivolv node does not have those limits. The specific
algorithms and number of floating points for each C will be evolved.
Some of the nodes should be able to activate in musical patterns,
but that is secondary to the evolved audio codes that directly generate
audio samples (like the networks of java objects of CodeSimian do).
The AI networks are for doing higher level thinking for planning
these evolved codes and creating other networks.
The evolved codes can become node algorithms.


Neural and Bayesian networks.
Evolved variations of bayesian networks
and neural networks
will do the slower evolution and general thinking.
Faster code will generate audio samples directly,
as CodeSimian does.
These faster codes will become parts of some of the network algorithms.


Human intuition.
Have you ever noticed how sometimes 30 seconds of some music
will stick in your head all day, depsite you only hearing the song once?
The mysteries of Brains are mostly unknown to us.
A little more accuracy in our thinking, coordinated by
computer generated music in realtime, is a good tool to learn why.
What if you could learn a phone number as easily
as that music was forced into your thoughts?

How will Audivolv work on the internet?
You will not notice it connecting to the internet,
but you will know because Audivolv will tell you.
You will just play music with the mouse which will customize
itself automatically to the way you like to move the mouse.
Audivolv will learn things like
how to cause one person to move his
mouse the same direction as another person moves
his mouse while using Audivolv on a different computer.
Using simple interactions like these,
complex interactions of unconscious Human thoughts will be organized
to create music and software, to improve Audivolv while Audivolv runs,
a self modifying artificial intelligence copied out of our own brains.

Learn to change music to control Humans hand on mouse
    Binary network of Funcs (each node has 2 childs).
    Each Func alone plays audio with mouse.
    Of many ways to move, control hand to move either of 2 specific ways.
    If not move either of those ways, weaken connection between those Funcs.
    If move 1 of those 2 ways, strengthen, and continue binary-search using mouse.

Evolve connectionist nodes/functions
    Node = network = constant-size array of variable-size arrays.
    In the same node, array sizes depend on eachother: CONSTANTRANGE, EQUAL, MULTIPLY, or POWER.
    Examples: Bayesian node. Audio node.
    HeapQueue is int[], int[], flo[].
    All arrays are always completely full.
    Manually choose recursion limit (depth 5), array and flo var limits (range -1 to 1).
    Evolve node types. Evolution chooses array types and their size equations.{
        Evolve Func that runs a loop, and what vars to: recurse, lookup in heapQueue, read/write flo{
            Evolve Func to be loop body, read/write constant number of flo (given by the loop Func).
    Play realtime audio, and thousands of times per second,
    change which 1-5 Funcs are currently playing, traversing and updating the network using the mouse.

Internet
    Has same behaviors as without internet, but smarter.
    Learn to copy 1 person's hand movements to other person's hand movements,
    while each person hears different music.
    Similar to the binary Func network, everyone does binary search on people.
    

REVERSE LINKS: ShortTermGoalsOfAudivolvDesign AUDIVOLV WEBSITES: http://audivolv.com | http://audivolv.sourceforge.net | http://audioevolve.com | http://instrumentsplaythemusicians.com