Theme: Causality

Let Me ‘Science’ These Statements by Carlos a Bit in Order to Demonstrate the Author Is Correct – But Offer a Minor Tweak.
RE: https://t.co/v9RTCz6AFB? by @IintuitMachine

1) RE: “- Computation is not an inherent category or natural kind, but arises from complex processes and interactions at multiple levels. It should be viewed as dynamically emergent rather than in static, formal terms.”–
And:
–“context-sensitive interaction with real-world environments”–
And:
–“Embodied Action”–

Disambiguation:
Memory is necessary for the organization of movement to obtain discounts on the pursuit of calories necessary for persistence.
Wayfinding consists of memory and recursion.
Thought consists of associative memory and wayfinding.
Consciousness consists of sufficient hierarchy of wayfinding memory to direct attention to recursively regulate wayfinding to find optimum choices.
Reason consists of disambiguation, comparison, and wayfinding.
Calculation consists of wayfinding by the transformation of inputs into outputs.
Computation refers to automatable transformation into outputs because of discrete and invariant categories.

SUMMARY
In this sense, with generalizing NNs trained by vast stores of language (categories of references, operations, and agreements(identity, consistency, correspondence) (=/!=)) we are evolving computation into calculation on our way to evolving through to wayfinding (branching iteration), in the hope that we can generalize that branching into consciousness of the choice between sets of predictions.

2) RE: – Computational systems are fundamentally embedded in and sustained by continual engagement with external environments, not isolated.–
And :
– The theory should focus on metaphysics and ontology first, not semantics. Meaning emerges from underlying processes.–
And :
— Computation unfolds dynamically over time through complex sequences of events, not as static symbolic manipulation. – Behavior emerges from underlying processes in context, not algorithms alone.’–
–And:
The goal of cognition is adaptive success through context-sensitive interaction with real-world environments.”–

ie: There is no closure possible without the informational content of reality, and tests of identity, consistency, operational possibility, and correspondence with it. This is why world models are necessary (Tesla’s AI) for language models (LLMs) to end arbitrariness, error, and hallucination.

The human brain creates the equivalent of a wireframe model underlying 3d computer games, including eye position, head position, torso position, limb position turning, speed, and relative direction, relative spatial location, and location relative to other locations because the primary purpose of the brain is to provide navigation (muscular action)through the world by disambiguation of senses consisting of nothing but pulses into a 3d model that the body size, speed, direction, and reach can move through. Everything else the brain does arises out of disambiguation, prediction, and wayfinding of bodily movement.
And while we are impressed with our progress, and while use of language to reverse-construct world models (the opposite of all of AI history) is a radical innovation we are not as close because the power law applies to this like every other problem. The last 10% is far harder than the previous 90 and so on, iteratively.

The human algorithm is trivial:
(a) “Maintain Homeostasis”, (b) “Get more of anything you can use to do that”, (c) ” With the greatest return, at the lowest cost, in the shortest time, with the greatest certainty, at the lowest risk”

The human capacity is equally trivial:
“Given a world model we can disambiguate with our senses, create episodic memories as indexes of combinations of objects, spaces, places, and locations continuously predict options for offense (get), rest, defense (run), and promote the optimum given state of homeostasis to attention, then recursively wayfind through predictions from episodic associations to achieve the desired result.”

The human ‘algorithm’ isn’t – it’s instrumental.
In that sense, the brain has ‘instruments’ that work together to use accumulated memories and continuous input, to produce the equivalent of a generalizable algorithm without the presence of any such algorithm other than the biomechanical pressures on excitement, auto association, prediction, attention, and recursion to restore homeostasis.

Tedious Simplicy Producing Complexity Through Quantity and Hierarchy.
In other words, human brains (all brains really) are trivially simple in general operation, just amazingly vast in adaptability given the use of many dendrites sparse patterns, dendrites, boutons, neurons, micro columns, macrocolums, amidst complexity of stimuli to produce a continuous stream of a nearly infinite set of permutations of potential outcomes operations that through wayfinding would make possible the restoration of homeostasis.

In other words the brain is mechanically simple:
(a) the brain just reducible to homeostasis inputs and hierarchical recursive memory.
(b) if you get the hardware right, there is no need for algorithms.
(c) consciousness, reason, rationalism, calculation, and computation will emerge in machine minds just as they have in the human mind by constraint on auto-association and increased burden on attention and working memory (frontal cortex).
(d) because the difference between each layer in the sequence of cognitive abilities,

Considerations on the Spread of “Mathiness”
IMO (and I’m over 60 and have been involved in the waves and the winters in AI – producing one of the first early AI’s (expert systems) for law), the problem is still the lack of neuromorphic hardware that allows adaptation in time, and this generation of pre-caculated transformers is a generation that will be wiped out by those that do not require pre-calculation.

I’m still not confident all the emphasis on mathematics necessary because the ‘mis-architecture’ of our chips might be as much of a problem as the einstein-bohr replatonization of mathematics has been for physics and the stagnation in physics because of it.

In other words, you’re working with math because you aren’t working with engineering. And you’re working with math because it’s CHEAPER and FASTER approximation of engineering. They way words and drawings are a cheaper faster way of imagining what might be.

Nature can’t do math without first evolving a brain that can. But it’s hella-difficult to defeat nature at evolutionary engineering. 😉

Cheers
Curt Doolittle
The Natural Law Institute
“The Logic of Everything”