Man-Machine Ontology for Machine Intelligence and AI/ML/LLM/AGI/ASI/Robotics
What is Man-Machine Ontology?
There is Universal/Big Ontology, Special/Domain Ontology, Human Ontology, Machine Ontology or Man-Machine Ontology
A?special domain ontology?is a set of models that comprehensively describe a given domain, like manufacturing, building structures, IoT systems, smart cities, energy grids, web content, and more.
Our human ontology is our mental model or cognitive schema of our world, call it worldviews, how we perceive and interpret reality. They integrate our tacit knowledge, experiences, beliefs, values, and assumptions, influencing our thoughts, emotions, behaviors, decision-making processes, or actions. Worldviews have the potential to evolve and deepen throughout our lives, influenced by human development.
Big/Generic Ontology is about all reality, the world, the universe, all existence and its content, as all entities and interactions
Man-Machine Universal Ontology is universal computational world ontology. It is about the Omniverse of Realities, physical, mental, social, digital or virtual.
No Machine Ontology, no Machine Intelligence and Learning, no real AI systems, no True Generative AI, no Real-World Large Language Models...
Machine Intelligence and Learning (MIL) as Real and True "AI"
Machine Intelligence and Learning (MIL) 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 infinite data to make predictions, recognize patterns, and make decisions based on new input data.
MIL as machine/technological/electronic/cybernetic/computing/digital/non-human intelligence is about intelligence in general,
its nature, elements, structures, functions, mechanisms,
and all possible implementations, be it natural intelligence, machine intelligence or alien intelligence.
So, real AI implies philosophy, general metaphysics or ontology, logics, semantics, mathematics, and statistics, cognitive sciences, linguistics, and psychology, data science and computer science.
In no ways, CS is the background knowledge, but one need to have ideas about algorithms, data models, information structures, computer architectures and programming languages.
To build real intelligent machines, we have to avoid a personification bias, attributing human characters and behavior, traits and emotions, thoughts or intentions to machines (non-human entities, like animals, objects, or abstract concepts).
True AI is the science and technology of machine intellect/intelligence, and in no ways the ability of digital computers or computer-controlled robots to mimic human bodies, brains, intelligence and behavior, as “developing systems endowed with the intellectual processes characteristic of humans, such as the ability to reason, discover meaning, generalize, or learn from past experience”.
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):
MIL & AI/ML/LLMs/AGI/Robotics Fundamentals = CWM = UCO [Ontological Multigraph Hypergraph Platform] + UCC [Causal Multigraph Hypergraph Engine] + USECS [Global World Knowledge Base]
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 is encoding the world of things, its domain and complex systems, as multi-graphic hypergraph networks of things, as Ontological Multigraph Hypergraph Platform, enabling World Embeddings, instead of word embeddings.
It is thus embracing conceptual/mathematical models of reality, theoretical structures, causal network diagrams, AI/ML models, all possible neural networks and graph neural network (GNN) architectures, all as special models, cases and topologies.
The Fundamentals of Universal Computational World Ontology
In?information and computer science, there is no universal ontology as studying the reality-data-intelligence-computation complex interactions.
It is simply an?ontology?which "encompasses a representation, formal naming, and definitions of the categories, properties, and relations between the concepts, data, or entities that pertain to one, many, or all?domains of discourse".
In terms of universal computational ontology, the highest genera, kinds or classes of things, T, the universal computational world ontology (UCWO) could be expressed as the world representation and reasoning reference framework:
UCWO = <W (T); D; I; Com; A> (1)
where W is the physical, biological, ecological, mental, social, digital or mathematical universe/world/reality, the universe of all possible universes, the world of all domains, the master category, the class of all classes and sets and subsets and elements; the universal class of all possible things, including nothing or nonbeing, nonentity as well, 0
T= <E, S, C, R; 0>
where E - the general classes of entities (substances, objects),
S - the general class of states (quantities/qualities, conditions or situations, states of affairs or facts);
C - the general class of changes, events, observations, phenomena, actions or processes;
R - the general class of relationships, links, associations, connections or interactions, like causality, space and time, similarity and analogy, or mathematical operations or functions, linear, power, quadratic, polynomial, rational, exponential, logarithmic, and sinusoidal;
D - the data universe of the World of Things, as Entity, State, Change and Relationship Data, data/information about reality, its domains and systems, data about entities (individuals, groups and objects, from quantum-mechanical to astronomical objects, as ephemerides), states (properties or attributes, quantities or qualities), changes *(variables, events, observations) or relationships (causality or interaction, time, space, etc.), respectively. Data are sets of values of qualitative or quantitative variables, belonging to a set of things. It may be in the form of numbers, letters, a set of characters, data signal of information, collected via observations, measurements or simulations. In data computing or data processing, data is represented by its type and structure, such as tabular data, data tree, a data graph, etc. Data are symbolized in the human mind, signified in natural language and its products, as documentation and literature and digital data, tokenized and embedded as numbers in the realm of Natural Language Processing (NLP), and LLMs, genAI and machine learning, and signaled via analogue or digital communication systems, natural or artificial.
I- Intelligence, the causal power of acquiring and understanding or comprehending world's data to transform into meaningful information or actionable knowledge, while involving learning, learning, abstracting, understanding, comprehension, all valid logic, inferences, reasoning, real-world problem-solving, which capacities and capabilities are dependent on its power to model and simulate and effectively interact with reality in terms of causal/logical algorithms and data types and structures;
Com is the class of computation functions, as mathematical equations problem solving or executing computing algorithms, arithmetic calculations, symbolic manipulations of all possible data types and structures, laws of nature, causal data patterns and regularities;
A - the class of data-processing, intelligent applications, as software, hardware or hybrid rational technologies, including AI, ML, LLMs, GenAI, AGI, ASI, or Robotics, ranging from the personal assistants and chatbots to space bots and autonomous intelligent spacecraft.
The critical question: "What are the fundamental capabilities and limitations of AI computers?" is replied the scope and scale of its computational or causal world models.
There are lots of confusion and misinterpretations what reality or world as a whole and its model mean, identified in different ways:
in philosophy and metaphysics, with "the sum or aggregate of all that is real or existent within the?universe, as opposed to that which is only?imaginary, nonexistent or nonactual".
in cosmology and physics, with "a totality of entities extending through space and time",
in psychology, with common sense, a collection of models of the world that can guide on what is likely, what is plausible, and what is impossible;
in predictive analytics or ML, with a neural network architecture for learning through observation and prediction, approximating human observation, learning, reasoning, planning, and acting … thinking;
in LLMs, as auto-regressive generative computational DNNs models of natural language data, text, images, video, etc.; "a general-purpose large-context multimodal autoregressive model, performing language, image, and video understanding and generation"
in human-mimicking intelligence, the predictive world model, the centerpiece of the Joint Embedding Predictive Architecture (JEPA) for autonomous human-like machine intelligence predicting plausible future states of the world.
The Fundamentals of Computational Causality
In computational ontology, computer science and engineering, AI and ML, reality is encoded as a causal world multi-graphic hypergraph network architecture used for learning, understanding through measurement, simulations, experiment, observation, inferences, prediction and interactions:
CWM = T x T = {(E, S, C, R) x (E, S, C, R)} (2)
Note it embraces a popular interpretation of causality "an influence by which one event, process, state, or object contributes to the production of another event, process, state, or object".
A causal world model is fully encoded with a multi-hypergraph which contains 4 types of nodes (vertices, points, or elements), as its order, 16 types of relationships to other nodes, as its size, with multiple/parallel each hyperedge (hyper-links, lines, arcs, arrows) and causal loops joining any number of elements for modeling multi-way group interactions among nodes.
Each node in the undirected multigraph hypergraph network with all possible causal loops is representing a causal factor or causal variable, as a cause or effect in four major categories:
as entities, substances, objects or agents, E,
states, quantities or qualities, S, as in with a causal state machine that defines states as causes and links each cause to a specific action,
changes, events, actions, activities, or processes, C,
领英推荐
relations, links, connections, or interactions, R.
Furthermore, a node may contain a collection of causes or even another causal graph. That means a causal model represented as a graph may store other causal sub-models in each node of the causal graph, thus representing complex causal structures.
The AI CWMs can reason over a single cause in the graph, a selected sub-graph, or the hypergraph itself, identifying real computational causality, C x C, as the cause-effect interrelationships or interactions of true causal variables, which are changes, events, actions, activities, or processes.
All real-world systems are modelled as causative multigraph hypergraphs, in which a causal link can connect any number of nodes, and a causal node might have any number of causal connections and loops.
In complex systems or interaction networks, one cause-variable has multiple effects-variables, and conversely, one effect may be caused by a multitude of causes, where causal emergence is the feedbacking effect of complex and nonlinear interactions between components.
Complex natural, living, cognitive, and artificial systems consist of diverse and heterogeneous elements/units/agents that interact through complex nonlinear CAUSAL relationships, modelled as causative multigraph hypergraphs with causal paths and feedback loops/circles/circuits, positive or negative, reinforcing or balancing, constructive or destructive.
In reality, the world is the largest environment ever, known as reality or existence or being or the universe, taken as a whole. It is the totality of all entities and interactions, of all matter and energy, of all space and time.
The AI CWM is representing its nature, structures, mechanisms, patterns, laws and principles by the Computational World Modeling Engine for Knowing, Learning, Inference and Interaction.
Universal Computational Ontology of Reality and its Contents
A Universal Computing Ontology serves as a foundation reference framework and a systematic, predictive, prescriptive descriptive, explanatory, causal integrating scheme, context, model, or structure of universal scope for:
Human Intellect and Knowledge, Individual and Collective
Philosophy, Metaphysics, Ontology, Epistemology, Logic and Semantics
Science, Technology, Engineering, Mathematics, Statistics, STEMS
the Arts and Humanities
a computational theory of cognition or the mind, Cognitive Computing, Cognitive Science
a computational model of human intelligence, AI, ML, Generative AI
natural language processing, generation, understanding, NLP/NLG/NLU
a computational theory of the human brain, ANNs, DL, Computational Neuroscience
a computational theory of information and computation, algorithms and data structures, automation and robotics, Computer Science and Engineering
a computational theory of data, Data Science and Engineering, Data Analytics
Computational Science or Scientific Computing
mathematical theory of circular causal control systems and processes, including in ecological, technological, biological, cognitive, social systems, Cybernetics, Systems Science
mathematical modeling of reality, Formal Foundation Ontologies, Domain Ontologies
empirical laws and scientific theories and models, from the Standard Model of elementary particles to general theories of political science and beyond
scientific modeling of reality, a physical, conceptual, or mathematical representation of real phenomena to explain and predict the behavior of real objects or systems in a variety of scientific disciplines, ranging from physics and chemistry to ecology and the Earth sciences.
It is a fundamental theory of reality that consist of a universal ontological system of intelligence so that from a limited number of prime principles, truths and categories, definitions and postulates, as assumptions and axioms, a whole body of empirically valid generalizations might be deduced in descending order of specificity, providing predictive inferences and causal explanations of changes, physical, biological, social, political, etc.
The Role of Reality Ontology in AI, ML, DL, LLMs, AGI and Robotics
Real Ontology serves as the general data backbone of real-world AI systems by providing a shared understanding of a world or its domains for humans and machines alike.
A universal ontology of reality, its entities, interactions, and properties, is a computer science and engineering construction to provide a universal reference framework for organising the world data, information or knowledge, including scientific knowledge and cultural knowledge, to be used by or through computer networks, as the internet/the web.
Entities and interactions, with their objects, systems and networks, data and knowledge, facts and laws, concepts and theories, models and paradigms, ideologies and movements, sciences and technologies, all is getting their meanings and roles within a unifying onto-semantic reference framework/context/environment.
As reified, the universal machine ontology would enable the development of computer systems that can know, learn, and reason about the world taking into account the interactions between entities in the actual world.
AI and ML and LLMs require large amounts of labelled training data for models to be accurately trained.
Ontologies, AI models and large language models (LLMs) complement each other perfectly for AI implementation because they address different aspects of knowledge representation and reasoning. When combined, they can lead to more robust, comprehensive, and contextually accurate AI solutions.
Unlike LLMs such as OpenAI’s GPT-4-5-x, a universal computational ontology would not depend on a huge amounts of training data, to reliably “know”, for example, that a a human being is an animal having a large braincase or that cats are warm-blooded animals who have a vertebrae, a 4 chambered heart and fur, which makes them mammals, while falling under the Kingdom Animalia, like humans, dogs or dolphins.
Given a sufficiently complete and detailed UCOR, a computer would “know” that “cat” as a subset of “mammal” which is in turn a subset of “animals” which are in the set of “living organisms” and so on.
By employing ontological models, AI/ML/DL/LLM systems can extract meaning from content and understand the context and semantics of data. This enables applications such as chatbots, language translation, recommendation engines, sentiment analysis, financial or healthcare systems to provide accurate and context-aware interactive responses.
In AI, real ontology acts as a language for expressing knowledge, allowing intelligent systems to comprehend and interpret data in a meaningful way. It not only helps in organizing information but also facilitates effective communication between different AI applications, making it a core element in building intelligent systems that can work collaboratively.
Furthermore, the UCOR in AI serves as a foundation for knowledge-based learning and and inference. By capturing the meanings, semantics and relationships between entities, it enables AI systems to perform complex tasks such as machine learning, deep understanding, natural language processing, meaningful search, automated inferencing, or intelligent interactions.
As the reality engine, the UCOR empowers AI models, as LLMs, to discover new knowledge by combining existing information and applying ontological/causal rules, leading to enhanced decision-making, predictive, prescriptive, descriptive and explanatory capabilities, following the formula:
AI = UCOR (World Knowledge, Learning, Inference, Interaction Engine) + AI/ML/DL Models + LLM databanks + ...
The universal computer ontology provides the basic descriptions of the meanings of data and algorithms used in specialized AI models and Large Language Models with local domain ontologies, knowledge graphs and local training datasets.
The universal computer ontology acts as a foundational framework to organize data, information and knowledge in advanced techno-scientific fields such as artificial intelligence, machine learning, information systems, data science, semantics, knowledge graphs, ontologies, deductive databases, large language models, etc.
In AI, the UCOR plays a Reality Engine role in world's data, information and knowledge representation, feature learning and inferencing, problem-solving and deciding, planning and interaction.
Optimizing logistics and transportation with a passion for excellence | Building Ecosystem for Logistics Industry | Analytics-driven Logistics
7 个月How can we ensure unbiased training data and models in AI/ML systems by utilizing Universal Computational World Ontology?
Chief AI Officer. Visionary technologist and lateral thinker driving market value in regulated, complex ecosystems.
7 个月Very good!! ??