A Theory of Reality for Machines and Humans: "the World Models" for Man-Machine Intelligence [AI/ML/DL/NNs/LLMs/GenAI/AGI/ASI]

[Who comprehends the world's knowledge owns the world. The human mind might know nothing, but understand everything. On the contrary, machines might know everything, but understand nothing, like LLMs trained with trillions tokens].

Our message is: we should go for true, real, scientific AI, leaving aside the mainstream AI folks creating the surrogates of human intelligence, led by the big tech fake AI $T club.

Study and develop “mental” models for true AI with causal intelligence and scientific world knowledge.

Such general "mental" world models for AI/ML/LLMs/AGI enable truly AI capacities and abilities: real intelligence, learning and understanding, explanation and prediction, creativity and discovery, inference and planning, interaction and adaptation.

The Iron Intelligence Rule: "no world mental models, no real intelligence and learning".

This is the fundamental reason why today's AI models as Machine Learning or Deep Learning or Neural Networks, Large Language Models or Generative AI or Artificial General Intelligence are deep unintelligent big data processing software/hardware systems, applications or tools.

What is Theory of Reality?

The theory of reality is an all-integrating framework or template to know and compute, understand and interpret the world as a whole, including the world around us. It provides a unifying structure through which humans or machines can analyze and synthesize or make sense of reality, its contents, and different aspects of reality.

Theories of reality can vary depending on the context, discipline or field of study, but they all aim to provide a systematic understanding of the nature of reality, its contents and dynamics,. These theories explain phenomena, discover effects, make predictions, and guide decisions, actions and interactions.

Today, no single theory has complete validity in all contexts, and different theories may have partial validity in different contexts or situations.

In the context of physics, for example, theories of reality may incorporate principles from both general relativity and quantum theory.

In the context of mathematics, theories of realities may incorporate the theory of category and set theory and graph theory.

In the general context of ontology, natural science, computer science, and mathematics, theory of reality incorporates the concepts and principles of formal ontology, physics, mathematics, graph theory and network theory.

Mathematics introduce formal abstract structures, the set with features, an operation, relation, metric, topology to study different aspects of reality, such as?measures,?algebraic structures?(groups,?fields, etc.),?topologies,?metric structures?(geometries),?orders,?graphs,?events,?equivalence relations,?differential structures, and?categories.

As the most generalized complex structure fitting to modeling all reality could be an undirected multi-graph hypergraph construct, in which an edge/link/relationship can join any number of nodes/vertices/entities/states/changes. and where a node can connect via multiple edges, all having multiple. parallel edges and loops.

If graphs can be used to model many types of relations and processes in physical, biological, social and information systems, an undirected complete multi-graph hypergraph of things in the world T could be applied to model all reality with its complex entities and interactions.

W = U = <T, R> (1)

where W is all reality or world as a whole or total universe, T is a universal set of things (vertices, nodes, points, elements), and R is universal set of hyperedges, lines, arcs, relationships, interactions, or cause-effect interrelationships.

General AI images, models, simulates or understands reality formally, symbolically and digitally operating within the multi-graphic hypergraph networks of things in the world, as entities, states, changes, processes or relations.

It could virtually "imagine" all the formulized world or its big parts depending on the supercomputing power and memory, unlike a limited human mental modelling: "The image of the world around us, which we carry in our head, is just a model. Nobody in his head imagines all the world, government or country. He has only selected concepts, and relationships between them, and uses those to represent the real system" (Forrester, 1971).

Reality as the Rational, Mathematical, Computable or Causal World

The world, reality or the universe is rational and mathematical and computable, measurable and observable, due to being essentially CAUSAL and REAL, in the first place

Metaphysical/Ontological/Causal/Objective/Scientific Reality lies behind, and within all the objective/material/physical/biological, mental/subjective/personal, social/intersubjective, digital/virtual, mathematical/logical/formal/ideal realities:

Ontological realism implies that the world and its things exist independently of how humans sense, perceive, experience, or think about it, and objective reality is the “gold standard” or scientific standard for what is real.

Different types of reality are as follows:

Physical reality with physical causes and effects, objects, systems, interactions, laws, rules, data, theories and models. machines and technologies. Physics is dealing with the world or universe as "[t]he totality of all matter and energy, space and time; all that is, has been, and will be". It is is about material reality, the natural science of matter and energy, “dealing with the structure of matter and the interactions between its fundamental constituents of the observable universe”.

Mental/Subjective reality with mental causes and effects, objects, systems, interactions, laws, rules, data, theories and models. machines and technologies. Many believe in a subjective, simulated, conscious reality, as "your reality is not my reality".

Social/ Cultural/Intersubjective reality with social causes and effects, objects, systems, interactions, laws, rules, data, theories and models. policies and institutions.

Ecological/Natural reality with environmental causes and effects, objects, systems, interactions, laws, rules, data, theories and models. machines and technologies.

Digital/Virtual reality with digital causes and effects, objects, systems, interactions, laws, rules, data and models. machines and technologies. It covers a whole spectrum of realities, from actual reality to virtual reality.

Formal/Mathematical/Logical reality with formal causes and effects, objects, systems, interactions, laws, rules, data theories and models. machines and technologies. As to a “mathematical universe hypothesis” anti-realism/idealism: all mathematical structures are physical structures, and that “only a mathematical world exists, with the finite, physical world being an illusion within it”.

Again, causal reality as a universal multi-graph hypergraph network of causes and effects might have zero, one, and more than one uncaused causes.

If there are no uncaused causes in causal reality, then causal reality involves an infinite regress of causes and effects: for each cause in causal reality, there is a causally prior cause in causal reality.

If there is one uncaused cause in causal reality, then causal reality has a unique initial cause, the cosmological singularity: like God or the big bang uncaused cause.

Note, the universe, the world or reality is not exhausted by physical reality…

In all, the world or reality or the universe is the totality of all things and interactions, objects or structures (actual and conceptual), changes, events (past and present), phenomena (observable or not) or processes.

It is what a universal world model ultimately attempts to describe or map, to model or simulate causal reality by means of transdisciplinary sciences, philosophical, natural, cognitive, social, technological, see the diagram above, aimed to codify all the knowledge in the world.

Who owns all the world's knowledge owns the world. This is why "Google" is a leading search engine; for Google's mission is to organize the world's information and make it universally accessible and useful.

It could be wise humanity or hyperintelligent machines or both of them. I advance for man-machine hyperintelligence...

Summing-Up

Here is the development formula of Real/True AI relying on the general theory of reality:

RAI = Machine Theory of Reality {Causal Multigraph Hypergraph Network} +

World Model (Intelligence, Learning, Inference, Interaction) Engine +

World Knowledge Engine {Global Knowledge Base, USECS + Large Language Models} +

Narrow AI Models (Symbolic AI rules, ML algorithms, Deep NNs, Chatbots) +

Software + Hardware (CMOS + GPU + Neuromorphic Processor, Quantum Processors)

>

(RAI Technology, RAI Supercomputing, RAI Internet, RAI Industries, RAI applications)

Resources

What Is Reality? AI Key to the World: Universal Ontology

Universal World Ontology for AI/ML/LLM/AGI/Robotics

We show that real intelligence machines and true learning systems come not from biased training data, statistical algorithms and numerical models, but from computational world models (CWM) relying on universal computational world ontology (UCWO) and computational causation (UCC), fueled by the Global Knowledge Base of USECS (Universal Standard Entity Classification System):

AI/ML/LLMs/AGI/Robotics Fundamentals = CWM = UCO [Ontological Multigraph Hypergraph Platform] + UCC [Causal Multigraph Hypergraph Engine] + USECS [Global World Knowledge Base]

[Real AI = Universal Computer Ontology + AI, ML, DL and LLMs]

We delve into Universal Computational World Ontology (UCWO) aka Reality Ontology, Universal Computer Ontology, Universal Mathematical Ontology or Global Ontology as the Causal World Modeling Engine for Learning, Inference, Understanding and Interaction.

A historical anecdote, we suggested the World Wide Web Consortium (W3C) to adopt our global ontology as a World's Information Reference Framework to enable the Semantic Web making the Internet data machine-readable. Instead, Tim Berners-Lee went for the RDF and OWL technologies ruining the whole enterprise.

Since history repeats itself, the same ending might happen with the big tech AI hype, if to ignore UCWO or Reality Ontology or Universal Computing Ontology or Formal Reality Ontology.

The UCWO, encoded as Ontological Multigraph Hypergraph Platform and enabling World Embeddings, instead of word embeddings, is embracing all possible neural networks, graph neural network (GNN) architectures as special topologies.

"Causal Fundamentalism": AI/ML/LLMs Fundamentals

Artificial Intelligence (AI) is built upon several key components that work together to enable intelligent machines, such as Data, ML Algorithms, Computing Power (CPU/GPU/TPU), and ML/DL models. AI models are mathematical representations of real-world processes created by training algorithms on data to make predictions, recognize patterns, and make decisions based on new input data.

We show that real intelligence and true learning come not from training data, statistical algorithms and models, but rather from computational world models (CWM) relying on universal mathematical ontology and computational universal causation.

We introduce "Causal Fundamentalism (CF)", as the fundamental principle of general computational ontology, AI and ML, as well as of science, engineering, technology and human practice.

USECS

Supplement on mental models

"A mental model is an internal representation of external reality: that is, a way of representing reality within one's mind.

Such models are hypothesized to play a major role in cognition, reasoning and decision-making.

The term for this concept was coined in 1943 by Kenneth Craik, who suggested that the mind constructs "small-scale models" of reality that it uses to anticipate events and that thinking itself is the manipulation of internal representations of the world.

Mental models can help shape behaviour, including approaches to solving problems and performing tasks. In psychology, the term mental models is sometimes used to refer to mental representations or mental simulation generally".