Developing artificial intelligence applications is hard work beyond buzz word chatter!
You have to train your mind to think like the
machine you're desiging.
Here's how to do it in a few steps.
17. 1 2 3 4 5
2
3
4
5
Numbers make up planes that in turn contain other numbers that represent
all of the concepts in MAPS. The whole system is in opposition to vectorial
architectures although vectors can be implemented in the model. This is
18. Euclidien Architecture : AGR representations are mostly
integers{0,1,2,3,4,5,6,7,8,9…} in a discrete mathematical model.
19. 1 2 3 4 5
2
3
4
5
1 2 3 4 5
2
3
4
5
1 2 3 4 5
2
3
4
5
These planes can contain any of the MAPS concepts numerically represented to reach a plausible
model, in any order and any type. There is no limit to the number of planes and the way they are
used to create the model. Please IMAGINE how these grids can apply to objects, their patterns,
properties and relationships right around you now.
20. 1 2 3 4 5
2
3
4
5
1 2 3 4 5
2
3
4
5
Genetic
native rules
such as 1 in
the plane on
the left
leads to 1, 2
and 3 in the
right plane
“Genetic algorithms are a class of probabilistic
algorithms which begin with a population of
randomly generated candidates and "evolve"
towards a solution by applying "genetic"
operators, modeled on genetic processes
occurring in nature. “ (Wikipedia). In the
AGR-MAPS system, the datasets are genetic
systems that set the course for the learning
algorithm. The probabilistic approach is the
plan of the plan process during which
evaluations of the level of data and the kind of
parameters to use are decided then AGR
algorithms are used as a transformation
function to verify the system and accept or
refute the result in a Bayes derived approach.
21.
22. 1 2 3 4 5
2
3
4
5
1 2 3 4 5
2
3
4
5
Genetic
native rules
such as 1 in
the plane on
the left
leads to 1, 2
and 3 in the
right plane
Toy problem : imagine all the ways a lawn mower could mow these surfaces (grdis, planes). Also how to figure out how to organize
the process without knowing where you’re going; you need a plan(datasets) and a calculator(your IMAGINATION ).
23.
24. Toy problem : imagine a organization that turns into chaos and then reacts into order
again with no hierarchy but through low level cooperation and a machine that identifies
25. Toy problem : imagine a organization that turns into chaos and then reacts into order
again with no hierarchy but through low level cooperation.