<under construction>

object_relationship_spaces_ai_ml

also see: https://github.com/lineality/definition_behavior_studies

also see: https://github.com/lineality/Online_Voting_Using_One_Time_Pads

Abridged Paper PDF in repo

Mini-Articles/excerpts Posted on Blog: https://medium.com/@GeoffreyGordonAshbrook

Articles

Sept 2023: Modeling Participant Architectural Learning in Five Trees Plus Mindstate
Sept 2023: The Vortex: Gender and AI's Scylla and Charybdis
Sept 2023: AI ALU Corpus Callosum
Sept 2023: Architectural Learning, Developmental-Landscape Hypervolumes, & Empirical Task-Trees: Participation With AI
Sept 2023: Subroutine Stacking Measure: AI Learning & General Project Participation
August 2023: AI Rules: Falstaff, Computer Science, and Natural Law
August 2023: Modularizing Problem Space for AI: Following a Wedge by Sight
August 2023: Questions as Objects in AI Object Handling
August 2023: Interpreting Hofstadter’s Gap: AI, Music, Math, Language
August 2023: AI Bodies & Brains: Solving A Problem
August 2023: Language, Analysis, Society, Compassion: AI, Biology, & Types of Intelligence
June 2023: Integrating STEM and Ethics for AI Alignment / Goal Alignment
May 2023: Scientific Method and Data Science Models
May 2023: AI Generalization’s Types & Testability
April 2023: Our AI Ancestors: Dumbledore’s Portrait and Ray Kurzweil’s Father
April 2023: Human-AI Interactions Study: World Chess Championships
April 2023: AI In A General Learning Gauntlet
April 2023: AGI's Culture Tools
April 2023: Better Tools To Plan AI

From v257: Abstract and Short Table of Contents

Definable Units of "Intelligence" for Evaluating AI Performance

Object Relationship Spaces for AI-ML: A Framework for 
Clearly Defined, STEM-Compatible, Project-Level, Functional Units of "Intelligence" 
For AI Design, Analysis, Performance, Architecture, and Operating Systems
Geoffrey Gordon Ashbrook, 2023.03-04




Abstract
There is a need for the use of well defined performance frameworks to describe the goals and skills/abilities of systems including AI. 

The overall agenda here is to move toward clearer communication and better definitions, including the pragmatic utilization of universal intersecting/interlocking areas.

This proposed object-relationship-space framework can be used for guiding project-specific system design, goal-setting, discussion, testing, analysis, reporting, regulation, documentation, etc. 
For more detail on what is meant by 'design': to manage and enable smaller or larger scale AI projects coordinating required abilities across internal and external components, including "symbolic" logistics and "sub-symbolic" training (including for AI-self-management), and whole operating-systems for AI. 

AI must be able to handle "objects" in the following interlocking contexts:
1. object-relationship-spaces
2. (internal/external) project-object-database (in a project-framework)
3. project-participants (in a project-framework & participation-space
such that 'objects' are defined as existing outside of the AI in an overall project context, and that so long as the AI effectively deals with these project-objects across these contexts including internal and external handling, it does not matter how the AI 'internally' handles objects. For example, alternative methods of internal handling/processing/management include:
- symbolic vs. sub-symbolic
- single or end-to-end vs. multiple or ensemble or hybrid
- parametric, nonparametric
- explainable vs. black-box
- higher dimensional vs. lower dimensional
- calculation vs. intuitive pattern recognition
- similar-to-h.sapiens, vs. not similar to h.sapiens
etc. 





This being a subset of a study of System & Definition behavior Studies, two useful sets of connections include Interconnecting/Intersecting Areas and Definitions, Frameworks, and Participation.

Interconnecting/Intersecting Areas:
A repeating theme, context, and agenda in this paper is to pragmatically leverage the interconnected functionality of clear definitions, STEM, projects, participation, coordination, system-fitness, positive values, and productivity.

To reiterate and state this as clearly and openly as possible, the context and agenda here is a project, best practice, positive-values, productivity, context. 


Definitions, Frameworks, and Participation
1. Discussion with undefined terms (for example specific abilities) can loop indefinitely regardless of the abilities of AI (at that time or in the case of changes over time in what AI can do). Undefined & under-defined goals, terms, and definitions tell us too little about what is needed, what the system can do, and if the system can do what is needed.

2. Telser Rule Loops: Where "AI" is undefined and every new development is dismissed as "not real AI," the failure to define "AI" tells us too little about what is needed, what the system can do, and if the system can do what is needed.

3. According to an Object-Relationship-Space framework in a project and participation context, AI-ML technology can as of March, 2023 join h.sapiens-humans as a participant in projects, with specific skills/abilities to handle specific project-objects, where projects, participants, and objects in object relations spaces, are clearly and functionally defined in a STEM context. This Object-Relationship-Space framework should define what is needed, what the system can do, and that the system can do what is needed. The details of how the practical context of 'project participation' partially overlaps with the vague context of 'intelligence' are likely significant.


Part one concerns a brief overview of the framework.

Part two concerns using the framework,
e.g. so you can construct your own well defined goals and tests for abilities of AI systems.

Part three concerns a discussion of the discussion of AI,
e.g. so you can critique statements in what you read about AI.

Part four concerns goals and agendas, background concepts and principles,
and future design factors.

Appendices include more examples and details.


Table of Contents: Brief

Abstract

A Narrative Introduction in Two Parts

Part 1: Framework in a Nutshell
The Object Relationship Space Framework
1.1  Example Object Relationship Space List
1.2  Many lists in One
1.3  Networked-AI Components

Part 2: Using The Framework
2.1  Examples: AI Skills, Comparing Three Chatbots
2.2  Examples: AI Skills Mapped to Object Spaces
2.3  Adding Levels, Adding Steps, Adding Objects
2.4  Hybrid/cross-model skills for AI
2.5  Heuristic & Pseudocode for AI Management with Object Spaces
2.6  Modularity, Scale, and AI-Component Networks
2.7  Model 'Explainability' as 'Explainability, 
      Reliability, and Security'
2.8  Mapping a general problems-space for AI & Mind
2.9  Object Relationship Based Testing 

Part 3: Discussing the Discussion of AI
3.1  Definitions of Terms 
3.2  Data Sources Discussing AI 
3.3  Examining Tests for AI 
3.4  What do we do with ChatGPT?

Part 4:  Goals, Background & Future
	4.1  Agenda & Goals
	4.2  Background Concepts and Principles
4.3  Future Design Factors

Appendices









Table of Contents: Detailed

Abstract

A Narrative Introduction in Two Parts:
Introduction Part 1. Chess in Blade Runner
Introduction Part 2. Defining AI Goals and 'Objects'

Part 1: Framework in a Nutshell -> The Object Relationship Space Framework
1.1  Example General Object Relationship Space List
1.2  Many lists in One
1.3  Networked-AI Components

Part 2: Using The Framework
2.1  Examples: AI Skills, Comparing Three Chatbots
2.2  Examples: AI Skills Mapped to Object Spaces
2.3  Adding Levels, Adding Steps, Adding Objects
2.4  Hybrid/cross-model skills for AI
2.5  Heuristic & Pseudocode for AI Management with Object Spaces
2.6  Modularity, Scale, and AI-Component Networks: and the need for a 
      literal or proverbial operating system.
2.7  Model 'Explainability' as 'Explainability, 
      Reliability, and Security'
2.8  General Mind-Space: Mapping a general problem-space for AI
2.9  Object Relationship Based Testing 

Part 3: Discussing the Discussion of AI
(In Summary)
3.1  Definitions of Terms
3.2  Data Sources Discussing AI
3.3  Examining Tests for AI
3.4  What do we do with ChatGPT?

(In Detail)
3.1  Definitions of Terms
	3.1.1  Terminology Issues 1: The tangled Semantics 
  of human ability. 
3.1.2  Terminology Issues 2: What has been defined or is 
  not-defined? 
	3.1.3  Terminology Issues 3: Navigate Jargon Pragmatically
	3.1.4  Terminology Issues 4: 
Fictional frames of reference are bad:
3.1.5  Terminology Issues 5: 
Problematic multiple meanings of unavoidable terms
	3.1.6  Terminology Issues 6: Beware Non-sequitur Conclusions
3.1.7 Discussions of Model "explanation"
	- Double standard between symbolic and subsymbolic
3.1.8  Reification
3.1.9  Definition Collapse: Maintain your definitions
3.1.10 Discussions of Model "explanation"
			- Double standard between symbolic and subsymbolic
3.1.10 Potemkin Villages and Telepathy-Tests
3.1.11  Terminology and Interpretation of Intent:
	- Azimov's Laws of Robotics
	- ELIZA the Psychotherapy AI
3.1.12 Terms that people cannot define while pretending they can
3.1.13 local context specific definitions
3.1.14  Negative Definitions
3.1.15  Indirect Definitions
	   'fail to disprove the nul hypothesis'
	   value function & meaning
3.1.16 Participation
3.1.17 Generalization
3.1.18 Controversial Topics
3.2  What To Read:	
		3.2.1  The Three-Legged Writing Stool
		3.2.2  History
		3.2.3  Interdisciplinary Area Recommendations
		3.2.4  Do AI Projects
		3.2.5  Book Recommendations
		3.2.6  Read classic Science Fiction: Back to Blade Runner again
3.3  Examining Tests for AI: 
		(Under Construction)
		3.3.1 - looking at winograd schemas
		3.3.2 - Sally Anne Tasks
	3.4  Influences on Model Architecture
	3.5  What do we do with Large Language Models & ChatGPT?

Part 4:  Goals, Background & Future:
	(In Summary)
4.1  Agenda & Goals
	4.2  Background Concepts and Principles
4.3  Future Design Factors

	(In Detail)
Introduction to Part 4
4.1  Agenda & Goals
		4.1.1  The goals and agenda here
		4.1.2  Defining your goals

	4.2  Background Concepts and Principles
4.2.1  "Intersecting/Interlocking Areas"
 		4.2.2  input output measures...or next section
		4.2.3  Higher Dimensional Frontier: Tensors & Matrices
		4.2.4  Projects & Project Context
	 	4.2.5  Instrumentalism and Realism
4.2.6  Big Other Areas
		4.2.7  Gamification
		4.2.8  Ambiguous Equivalence

	4.3  Future Design Factors
		4.3.1  Biology
			4.3.1.1  Integration with biological systems
			4.3.1.2  Use or imitation (of biological functions)
			4.3.1.3  Compare and contrast for study and understanding
			4.3.1.4  Highlighting known areas of development
			4.3.1.5  Highlighting still not well understood areas
			4.3.1.6  Highlighting predictable problems
4.3.1.7  The science of sleep
4.3.1.8  The science of memory
4.3.1.9  Non-chordata "intelligence" & decision making
4.3.1.10  Science of Mind/Cognition/Consciousness
4.3.1.11  Science of Entheogens
4.3.1.12  Science of Mindfulness
4.3.1.13  Bio-Nano-Coded AI
4.3.1.14  AI in synthetic organisms for terraforming
4.3.1.15  DNA/RNA based digital information interface
4.3.1.16  The Dragon Project (hybrid synthetic chimeras)
		4.3.2  Cybersecurity and AI
		4.3.3  Quantum information theory & under-the-hood optimizations
		4.3.4  "Generalization" vs. Deployment
		4.3.5  Nanotech
		4.3.6  Understanding Exponential Elbows
		4.3.7  'Complexity' Nonlinearity Dynamical and Systems Sciences
4.3.8  Ethics, Projects, Best Practice & STEM 
  STEM, Ethics & Mindfulness
		4.3.9  Projects: Agile
4.3.10  AI and Code Testing
		4.3.11  The Long Term Memory Storage Problem
		4.3.12  The challenge of orientation and navigation in mind-space
		4.3.13  Human machine interactions, biology machine integration
4.3.14  Project-Context Decision-Making Involving Participants 
   and Components
		4.3.15  Question Space
		4.3.16  Self-Awareness Apace
		4.3.17  Analogies
		4.3.18  System Epidemiology
		4.3.19  The Cambrian Midway Point
		4.3.20  Parent-Child Policy Decision
		4.3.21	  Culture & AI
4.3.22  Kasparov Event Horizon for Object Perception & Handling
4.3.23	  Scientific Method and Data Science Models
4.3.24  AI Bodies and Brains
4.3.25  Interpreting Hofstadter's Gap: AI, Music, Math, Language
4.3.26  Modularizing Problem Space for AI
4.3.27  Questions as Objects in AI Object Handling
4.3.28  AI, Biology, & Types of Intelligence
4.3.29  CS, STEM, law, natural-law
4.3.30  Architectural Learning,  Developmental-Landscape
 	   Hypervolumes, &  Empirical Task-Trees: Participation
4.3.31  Subroutine Stacking Measure: AI Learning
4.3.32  AI ALU Corpus Callosum 
4.3.33  The Vortex: Gender and AI's Scylla and Charybdis
4.3.34  Modeling Participant Architectural Learning 
         in Five Trees Plus Mindstate




Appendices: 
Appendix #: Recommended Reading & Extended Reading
Appendix #: Expanded Introduction
Appendix #: Expanded Part 1
Appendix #: Expanded Part 2
Appendix #: Expanded Part 3
Appendix #: Expanded Part 4