070-Basic Model: outline: how to transition from rep theory to ai problems, and then to ai making

The development of a basic model is a collection of the ideas and concepts that I worked through as necessary to construct a workable machine consciousness.

Rep theory (which is basically done - make sure the change model is explained)

make sure that duality is described.  it's important for doing any kind of grouping and ungrouping  for doing synthesis and analysis. duality is a key observational fact about representation.  duality is the key feature of mirroring

have to lay out all the arbitrariness arguments.  the profound arbitrariness of representation is key to similarity and difference, to representation itself, but especially to meaning.  it's why a  change in perspective can change meaning.

make sure to describe representation and representation making takes as two forms   (a,b) -> c  or (a,b) ; C     the one is a transformation.  the other is information.  they both are representation and representation functions and making.   but they serve different purposes.   (in chemistry, they are both transformations  (a,b) -> c   and (a,b) -> d   but d is an informational transformation and c is a functional transformation.  (some other d related process likely reverts d to (a,b)  or there is some catalyst process with a protein to produce information.    function and awareness.

IT'S REPRESENTATION ALL THE WAY DOWN!

representation issues:

• colors are arbitrary
• arbitrariness is qualia.
• arbitrariness is not randomness, but it can be.   randomness can be both arbitrary and not arbitrary.
• demonstrate the arbitrariness is the rule of the day.  we think everything it logical and follows rules, but that belief itself is arbitrary.  it's an arbitrary preference.   survival is arbitrary.  there is not "rule" for things in the universe, arbitrariness rules the day.
• categories are illusions like states.  we recognize categories, but categories do not actually exist when you examine them.  like sunsets or unicorns.  - categories are not extrinsic, but intrinsic groupings of association. (it's why they can be paradoxical)
• selection without meaning, without following a rule.   this also ties into stigmergy.
• correlation and co-occurrence.  causation and co-occurrence
• the importance and prevalence of mirroring.  both in representation and as neural structures and as chemical networks

problems for making an ai
this is all the stuff i had to go over and why it was useful or didn't work.
• petri nets  to many things it can't do.
• mathematics/algorithms
• fails for where is the 2
• fails for where are the algorithms stored (the reps are in men, not in the machine)
• fails because math cannot generate representations (can only categorize existing representations again, reps in men
• basic computation model / automata
• fails because representations come out of nowhere.  gives us a rule: conservation of bytes.  if data comes out of nowhere, then it must be generated by a rule.  again, where is the rule?   and that leads us down the rabbit hole of homunculi or men.
• the rule set specifically.
• algorithms all fail because:
• where is the rule?
• discovering an algorithm, then implementing it as a computation are both representational problems, so there is a rule set for that behavior, where is it?
• if a a computer implements an algorithm as a computation, it doesn't know what the algorithm is.  The computation is not the algorithm.  the computer is not conscious of it's own computations as representations, as algorithms.
• neural nets fail (see aom entry)
• Cellular automata.
• many interesting features, but we are stuck with the rule problem
• cellular automata are not moving, it's a static system.  fluid cellular automata
• halting conditions.
what is going on in biology?
• ants and ant algorithms.  The work for the problem set, but ant's are not handed a pre-programmed problem set.
• how slime molds find their way through a maze?
• how do bacteria find food and avoid danger?  is there an algorithm there?  no, there clearly is not algorithm.
• neuron's do not link up because of an algorithm  neuron's do not fire because of an algorithm, the fire because of local conditions.
• multicellular organism that move.  motion is key feature of representation making by nervous systems. both to move, and to not move.
• why?  because the moving body must survive.  it must move to eat and m one to avoid being eaten.  similar to bacteria, but more complex.
• conceptual embodiment is a by-product of moving organisms  both to go somewhere and get out of the way.
• movement prediction offers obvious advantages over non-predicting movements.  this gives us all kinds of sensation and the central nervous system itself.
• ants give us stigmergy
• stigmergy is arbitrary
• stigmergy in ants looks like an actor kind of model.
• stigmergy also looks like a cellular automata model.
• a model where the rules are elicited by some cells, and propagated to other cells.
• this rule selection does not get us out of the where is the rule problem, but it does get us to moving rules around
• stigmergy extends to cells.
• how to explain the behavior single celled organisms.
• systems biology.
• a brief overview covering basic layouts in text book
• models of cell interiors using fluid cellular automata, where the cell does move around.  still have the rules problem.
• the pattern of beetle shells is both systems biology, but it is also cellular automata.
• explain the decay of systems biology of a signal and the transformation of cellular automata to make patterns.
• the cellular automata are fluid in systems biology and the transformations are conserving (bit complete)
• saturation, desaturation
• over saturation is a halting condition
• the molecule and the change of interacting molecules
• molecules look like representations  (x,y) -> z  this is what actually happens.  we talk about it differently, but it's only our view that says a molecule takes on small bits and remains the same,  the reality is the small bit is added on and remains the same too.
• all molecular interactions are two way.
• proteins/catalysts.
• proteins must be destroyed and created
• this has a corollary to cellular automata.  if we let an automata run in most rules, it becomes stale and stagnant.
• structures of cells, how do they form, no rules, it's stigmergy and systems biology.  but it is also conservation of mass.
• conservation of mass, there is no magic creation and destruction as we find in cellular automata.
• the cell must survive.  it must contain it's functions.  embodiment.
• a cell wall and a cell interior.  they work differently.  duality and embodiment
• how to make structures and have a fluid space?
• structures are like programs, monolithic programs.
• fluid spaces with proteins, molecules etc are like super tiny programs, single function, single step programs/automata
• the cell must make representations.  embodiment again.  the rapper,  how does it make representations?
• it must have representation making as part of it's body.  that is where the rules are.  in the body.
• the body is the rules.
• the function of representation making must be embodied.
• this is exactly what we see in cells. the proteins, the dna, the ribosome, all of these things are representation making functions.  by having all these functions,  by having all these functions inside it, the cell embodies representation making
• the cell was is a representation making function too.  what goes in and out, these are representations.
• neural cells signal and that is a representation.  how do we know? because we can copy it fairly closely with another representation - the neural network/action potential model.  the two are doing the same kind of representation.  that is how we know the neuron's work the way they do
• the cell is more complex than that one action potential signaling model.
• the network of cells embody representations.  the structurer of representations is the structure of cells and their signaling.
• all of the different kinds of neurons and other cells matter.  motive action - stigmergy
• dopaminergic neurons etc  meaning is in chemistry
• meaning is in chemistry as demonstrated by drugs.  hallucinogens open up meaning and perspective. it shows the structure and function of the neurons is what matters to experience, to representations.
• imagination and other features of the basic model
• How to construct cells in code?  that is the big problem:
• stigmergy
• embodiment
• bit completeness/conservation of bytes (no magic actions)
• not static, does not decay into a cellular automata stasis
• no encoded rules

Basic Model:
• the model must be representation all the way down (there is no magic)
• resolution of mind/body duality - idea/physic duality is a magic argument
• simulation is the key problem of cartesian and physicalist duality
• explanatory solution to the causal interaction between representational ideas and non-representational material
• arbitrariness
• embodiment
• the hard problem disappears in the face of structured non-representational objects in a system making representational structures. structure = representation
• all ideas must manifest as neurological (later computational) structures.  otherwise, where are they?  how the causal interaction happens
• network structures and behavior
• why cells, and how do cells connect to each other to create system of representations, representational happenings (simulation), and develop representations.

• accounting for stigmergy
• systems biology is the model for stigmergic behavior in the cell.
• protein as a representational base, a minimal repper.  (a,b) -> c  the chemistry is explicitly a non-representational process
• conservation of mass / conservation of bytes - or no magic in the non-representational process
• which computational approaches will work?   anything algorithmic fails categorically, it is neither representation all the way down, but has no non-representational base - algorithms stick us with mind/body  rep/non-rep duality
• no-state (state is illusion)
• no rules - where are the rules?  There are no newtonian "laws", there are only newtonian descriptions. how does the "law" effect the physical world?  (the causal problem inverted)  laws are "magic"    the physics must instantiate representations.
• imagination and other features of the basic model
• How to construct cells in code?  that is the big problem:
• fluid chemistry/cellular automata,  membrane chemistry/cellular automata - resolving this problem
• bit completeness/conservation of bytes (no magic actions)
• not static, does not decay into a cellular automata stasis
• no encoded rules
• computational atoms
• computations care about outcomes - representational at the bottom
previous next