Deploy from jjohare/logseq @ 7364d4dad5da93f34e6c59826e5a3e55095d5ec9 jjohare/logseq@7364d4d

Author: jjohare

api/pages/pages/3D Rendering Engine.json

@@ -12,18 +12,18 @@
     "Avatar"
   ],
   "wiki_links": [
-    "Computer Graphics",
     "Virtual Reality",
-    "Unreal Engine",
-    "Game Development",
-    "XR Applications",
-    "Blender",
     "Metaverse",
-    "Digital Twin",
-    "Real-time Visualisation",
+    "Computer Graphics",
     "Unity",
     "Augmented Reality",
-    "Immersive Experiences"
+    "Digital Twin",
+    "XR Applications",
+    "Unreal Engine",
+    "Game Development",
+    "Immersive Experiences",
+    "Real-time Visualisation",
+    "Blender"
   ],
   "ontology": {
     "term_id": "NGM-7009",

api/pages/pages/A-Star Algorithm.json

@@ -7,64 +7,64 @@
     "RB-1018-dijkstra-algorithm"
   ],
   "wiki_links": [
-    "Route Planning",
-    "Optimal Path Discovery",
-    "Hash Table",
-    "Navigation",
-    "Informed Search",
+    "Google Maps Platform",
+    "Informed Search Strategy",
+    "Robotics Control",
+    "Distance Metric",
+    "Admissible Heuristic",
+    "Video Game AI",
+    "Data Structures",
+    "Motion Planning",
+    "Graph Data Structure",
+    "Russell & Norvig Artificial Intelligence Modern Approach",
+    "AlgorithmicFramework",
+    "Network Routing",
+    "Robotics Industry Association Automation Statistics",
     "Heuristic Methods",
-    "Closed Set",
-    "Node Expansion",
+    "AlgorithmicLayer",
+    "Cormen Introduction to Algorithms",
+    "Dijkstra's Algorithm",
+    "Graph Algorithms",
     "Path Reconstruction",
+    "Informed Search",
+    "Closed Set",
+    "Autonomous Navigation",
+    "ApplicationLayer",
     "Uniform Cost Search",
-    "Heuristic Evaluation",
-    "Distance Heuristics",
-    "Logistics Optimization",
+    "International Planning Competition",
+    "Algorithmic Efficiency",
+    "ComputationAndIntelligenceDomain",
+    "Node Expansion",
     "Hart, Nilsson, Raphael 1968 Formal Basis for Heuristic Determination",
-    "Autonomous Navigation",
+    "Search Algorithms",
+    "Navigation",
+    "Priority Queue",
     "Greedy Best-First Search",
-    "Motion Planning",
-    "Dijkstra's Algorithm",
-    "Evaluation Function",
+    "Unity Technologies NavMesh Documentation",
+    "Best-First Search",
     "Heuristic Search",
-    "Breadth-First Search",
-    "Russell & Norvig Artificial Intelligence Modern Approach",
-    "Graph Algorithms",
-    "Cormen Introduction to Algorithms",
-    "Robotics Industry Association Automation Statistics",
-    "ComputationAndIntelligenceDomain",
-    "AlgorithmicFramework",
-    "Google Maps Platform",
-    "Pathfinding",
-    "Informed Search Strategy",
-    "Priority Queue Data Structure",
-    "Goal-Directed Search",
-    "Distance Metric",
     "AI-GroundedDomain",
-    "Search Algorithms",
-    "Optimization Algorithms",
-    "AlgorithmicLayer",
+    "Open Set",
+    "Optimal Pathfinding",
+    "Heuristic Evaluation",
+    "Logistics Optimization",
+    "Breadth-First Search",
     "Heuristic Function",
-    "Graph Representation",
-    "International Planning Competition",
-    "ApplicationLayer",
-    "Priority Queue",
-    "Best-First Search",
-    "Bidirectional Search",
+    "Pathfinding",
     "Cost Function",
-    "Admissible Heuristic",
+    "Iterative Deepening A*",
+    "Bidirectional Search",
+    "Hash Table",
+    "Graph Representation",
+    "Distance Heuristics",
+    "Optimal Path Discovery",
     "Graph Theory",
-    "Unity Technologies NavMesh Documentation",
-    "Optimal Pathfinding",
-    "Robotics Control",
+    "Optimization Algorithms",
+    "Priority Queue Data Structure",
     "ROS Navigation Stack",
-    "Iterative Deepening A*",
-    "Video Game AI",
-    "Network Routing",
-    "Graph Data Structure",
-    "Algorithmic Efficiency",
-    "Data Structures",
-    "Open Set"
+    "Evaluation Function",
+    "Route Planning",
+    "Goal-Directed Search"
   ],
   "ontology": {
     "term_id": "AI-1004",

api/pages/pages/AI Applications.json

@@ -5,11 +5,11 @@
   "backlinks": [],
   "wiki_links": [
     "Medical AI",
+    "Artificial Intelligence",
     "AI Assistant",
     "Autonomous Vehicles",
     "AI Deployment",
-    "Industrial AI",
-    "Artificial Intelligence"
+    "Industrial AI"
   ],
   "ontology": {
     "term_id": "AI-0604",

api/pages/pages/AI Concept.json

@@ -6,11 +6,11 @@
     "ModelProperty"
   ],
   "wiki_links": [
-    "owl:Thing",
-    "Machine Learning",
     "Knowledge Representation",
+    "owl:Thing",
     "Artificial Intelligence",
-    "Cognitive Architecture"
+    "Cognitive Architecture",
+    "Machine Learning"
   ],
   "ontology": {
     "term_id": "AI-9002",

api/pages/pages/AI Framework.json

@@ -9,11 +9,11 @@
     "Intelligent Virtual Entity"
   ],
   "wiki_links": [
-    "Neural Network Library",
+    "Model Optimization",
     "owl:Thing",
     "AutoML",
     "Deep Learning Framework",
-    "Model Optimization"
+    "Neural Network Library"
   ],
   "ontology": {
     "term_id": "AI-9005",

api/pages/pages/AI Hardware.json

@@ -14,15 +14,15 @@
     "Tensor Processing Unit"
   ],
   "wiki_links": [
+    "Computer Hardware",
+    "Artificial Intelligence",
+    "Deep Learning",
+    "Large Language Models",
     "AI Infrastructure",
     "Neural Networks",
+    "AI Training",
     "High-Performance Computing",
-    "Deep Learning",
-    "Computer Hardware",
-    "Large Language Models",
-    "Machine Learning",
-    "Artificial Intelligence",
-    "AI Training"
+    "Machine Learning"
   ],
   "ontology": {
     "term_id": "AI-7020",

api/pages/pages/AI Infrastructure.json

@@ -7,10 +7,10 @@
   ],
   "wiki_links": [
     "Edge AI",
-    "GPU Computing",
-    "AI Deployment",
     "Artificial Intelligence",
-    "Cloud AI"
+    "GPU Computing",
+    "Cloud AI",
+    "AI Deployment"
   ],
   "ontology": {
     "term_id": "AI-0603",

api/pages/pages/AI Safety.json

@@ -10,12 +10,12 @@
     "AI Risks"
   ],
   "wiki_links": [
-    "Interpretability",
-    "AI Alignment",
-    "Value Alignment",
     "AI Governance",
     "Artificial Intelligence",
-    "Adversarial Robustness"
+    "Value Alignment",
+    "Interpretability",
+    "Adversarial Robustness",
+    "AI Alignment"
   ],
   "ontology": {
     "term_id": "AI-4009",

api/pages/pages/AI-0376-algorithmic-accountability.json

@@ -4,11 +4,11 @@
   "content": "- ### OntologyBlock\n  id:: algorithmic-accountability-ontology\n  collapsed:: true\n\n  - **Identification**\n    - public-access:: true\n    - ontology:: true\n    - term-id:: AI-0376\n    - preferred-term:: Algorithmic Accountability\n    - source-domain:: ai\n    - status:: complete\n    - version:: 1.0\n    - last-updated:: 2025-10-28\n\n  - **Definition**\n    - definition:: Algorithmic Accountability is a responsibility framework that ensures AI systems and their developers are answerable for the decisions, outcomes, and impacts produced by algorithmic processes, including mechanisms for redress, transparency, and oversight.\n    - maturity:: mature\n    - source:: [[IEEE P2863]]\n    - authority-score:: 0.95\n\n  - **Semantic Classification**\n    - owl:class:: ai:AlgorithmicAccountability\n    - owl:role:: Process\n    - owl:inferred-class:: ai:VirtualProcess\n    - belongsToDomain:: [[AIEthicsDomain]]\n    - implementedInLayer:: [[ConceptualLayer]]\n\n  - #### Relationships\n    id:: algorithmic-accountability-relationships\n    - is-subclass-of:: [[AIGovernancePrinciple]], [[EthicalFramework]], [[RegulatoryCompliance]]\n\n  - #### OWL Axioms\n    id:: algorithmic-accountability-owl-axioms\n    collapsed:: true\n    - ```clojure\n      ;; Algorithmic Accountability Ontology (OWL Functional Syntax)\n;; Term ID: AI-0376\n;; Domain: AIEthicsDomain | Layer: ConceptualLayer\n\n(Declaration (Class :AlgorithmicAccountability))\n\n;; Core Classification\n(SubClassOf :AlgorithmicAccountability :AIGovernancePrinciple)\n(SubClassOf :AlgorithmicAccountability :EthicalFramework)\n(SubClassOf :AlgorithmicAccountability :RegulatoryCompliance)\n\n;; Annotations\n(AnnotationAssertion rdfs:label :AlgorithmicAccountability \"Algorithmic Accountability\"@en)\n(AnnotationAssertion rdfs:comment :AlgorithmicAccountability\n  \"Responsibility framework ensuring AI systems and developers are answerable for algorithmic decisions, outcomes, and impacts through mechanisms for redress, transparency, and oversight\"@en)\n(AnnotationAssertion :isoReference :AlgorithmicAccountability \"IEEE P2863-2021\")\n(AnnotationAssertion :authorityScore :AlgorithmicAccountability \"0.95\"^^xsd:float)\n(AnnotationAssertion :priorityLevel :AlgorithmicAccountability \"4\"^^xsd:integer)\n\n;; Object Properties - Accountability Mechanisms\n(SubClassOf :AlgorithmicAccountability\n  (ObjectSomeValuesFrom :requiresAuditTrail :TraceabilityMechanism))\n(SubClassOf :AlgorithmicAccountability\n  (ObjectSomeValuesFrom :enablesRedress :RedressProcedure))\n(SubClassOf :AlgorithmicAccountability\n  (ObjectSomeValuesFrom :assignsResponsibility :AccountableParty))\n(SubClassOf :AlgorithmicAccountability\n  (ObjectSomeValuesFrom :implementsOversight :GovernanceStructure))\n(SubClassOf :AlgorithmicAccountability\n  (ObjectSomeValuesFrom :assessesImpact :AlgorithmicImpactAssessment))\n(SubClassOf :AlgorithmicAccountability\n  (ObjectSomeValuesFrom :verifiesCompliance :ComplianceAudit))\n\n;; Object Properties - Governance Integration\n(SubClassOf :AlgorithmicAccountability\n  (ObjectSomeValuesFrom :alignsWith :OECDPrinciple))\n(SubClassOf :AlgorithmicAccountability\n  (ObjectSomeValuesFrom :satisfies :EUAIActRequirement))\n(SubClassOf :AlgorithmicAccountability\n  (ObjectSomeValuesFrom :implements :IEEEStandard))\n\n;; Data Properties\n(DataPropertyAssertion :hasAccountabilityLevel :AlgorithmicAccountability \"comprehensive\"^^xsd:string)\n(DataPropertyAssertion :requiresHumanOversight :AlgorithmicAccountability \"true\"^^xsd:boolean)\n(DataPropertyAssertion :enablesContestability :AlgorithmicAccountability \"true\"^^xsd:boolean)\n(DataPropertyAssertion :mandatesDocumentation :AlgorithmicAccountability \"true\"^^xsd:boolean)\n\n;; Property Declarations - Accountability Relations\n(Declaration (ObjectProperty :requiresAuditTrail))\n(ObjectPropertyDomain :requiresAuditTrail :AlgorithmicAccountability)\n(ObjectPropertyRange :requiresAuditTrail :TraceabilityMechanism)\n\n(Declaration (ObjectProperty :enablesRedress))\n(ObjectPropertyDomain :enablesRedress :AlgorithmicAccountability)\n(ObjectPropertyRange :enablesRedress :RedressProcedure)\n\n(Declaration (ObjectProperty :assignsResponsibility))\n(ObjectPropertyDomain :assignsResponsibility :AlgorithmicAccountability)\n(ObjectPropertyRange :assignsResponsibility :AccountableParty)\n\n(Declaration (ObjectProperty :implementsOversight))\n(ObjectPropertyDomain :implementsOversight :AlgorithmicAccountability)\n(ObjectPropertyRange :implementsOversight :GovernanceStructure)\n\n(Declaration (ObjectProperty :assessesImpact))\n(ObjectPropertyDomain :assessesImpact :AlgorithmicAccountability)\n(ObjectPropertyRange :assessesImpact :AlgorithmicImpactAssessment)\n\n;; Data Property Declarations\n(Declaration (DataProperty :hasAccountabilityLevel))\n(DataPropertyDomain :hasAccountabilityLevel :AlgorithmicAccountability)\n(DataPropertyRange :hasAccountabilityLevel xsd:string)\n\n(Declaration (DataProperty :requiresHumanOversight))\n(DataPropertyDomain :requiresHumanOversight :AlgorithmicAccountability)\n(DataPropertyRange :requiresHumanOversight xsd:boolean)\n\n(Declaration (DataProperty :enablesContestability))\n(DataPropertyDomain :enablesContestability :AlgorithmicAccountability)\n(DataPropertyRange :enablesContestability xsd:boolean)\n      ```\n\n- ## About Algorithmic Accountability\n  id:: algorithmic-accountability-about\n\n  - Algorithmic Accountability is a responsibility framework that ensures AI systems and their developers are answerable for the decisions, outcomes, and impacts produced by algorithmic processes, including mechanisms for redress, transparency, and oversight.\n  -\n    - ### Implementation Patterns\n  - ### Pattern 1: Accountability Registry System\n    ```python\n    from dataclasses import dataclass\n    from datetime import datetime\n    from typing import List, Optional, Dict\n    from enum import Enum\n  -\n    class AccountabilityRole(Enum):\n        \"\"\"Roles with accountability for AI system stages\"\"\"\n        DATA_PROVIDER = \"data_provider\"\n        MODEL_DEVELOPER = \"model_developer\"\n        SYSTEM_DEPLOYER = \"system_deployer\"\n        OVERSIGHT_AUTHORITY = \"oversight_authority\"\n        REDRESS_HANDLER = \"redress_handler\"\n  -\n    @dataclass\n    class AccountableParty:\n        \"\"\"Individual or organisation responsible for AI system aspect\"\"\"\n        party_id: str\n        name: str\n        role: AccountabilityRole\n        contact_information: Dict[str, str]\n        responsibility_scope: str\n        liability_coverage: Optional[str] = None\n  -\n    @dataclass\n    class AlgorithmicDecision:\n        \"\"\"Record of individual algorithmic decision for audit trail\"\"\"\n        decision_id: str\n        timestamp: datetime\n        input_data: Dict\n        output_decision: Dict\n        model_version: str\n        confidence_score: float\n        accountable_party: str\n        audit_metadata: Dict\n  -\n    class AccountabilityFramework:\n        \"\"\"Implementation of comprehensive accountability framework\"\"\"\n  -\n        def __init__(self):\n            self.accountability_registry: Dict[str, AccountableParty] = {}\n            self.decision_audit_trail: List[AlgorithmicDecision] = []\n            self.redress_procedures: Dict[str, callable] = {}\n  -\n        def register_accountable_party(self, party: AccountableParty) -> None:\n            \"\"\"Register individual or organisation with accountability\"\"\"\n            self.accountability_registry[party.party_id] = party\n  -\n        def log_decision(self, decision: AlgorithmicDecision) -> None:\n            \"\"\"Record algorithmic decision in audit trail for traceability\"\"\"\n            if decision.accountable_party not in self.accountability_registry:\n                raise ValueError(f\"Accountable party {decision.accountable_party} not registered\")\n            self.decision_audit_trail.append(decision)\n  -\n        def initiate_redress(self, decision_id: str, contestation_reason: str) -> Dict:\n            \"\"\"Enable individuals to contest algorithmic decisions\"\"\"\n            decision = next((d for d in self.decision_audit_trail if d.decision_id == decision_id), None)\n            if not decision:\n                return {\"status\": \"error\", \"message\": \"Decision not found in audit trail\"}\n  -\n            accountable_party = self.accountability_registry[decision.accountable_party]\n  -\n            return {\n                \"status\": \"redress_initiated\",\n                \"decision_id\": decision_id,\n                \"accountable_party\": accountable_party.name,\n                \"contact\": accountable_party.contact_information,\n                \"contestation_reason\": contestation_reason,\n                \"review_timeline\": \"30 days\"\n            }\n  -\n        def generate_accountability_report(self) -> Dict:\n            \"\"\"Produce comprehensive accountability documentation\"\"\"\n            return {\n                \"total_decisions\": len(self.decision_audit_trail),\n                \"accountable_parties\": len(self.accountability_registry),\n                \"decisions_by_party\": self._aggregate_decisions_by_party(),\n                \"redress_statistics\": self._calculate_redress_metrics(),\n                \"audit_completeness\": self._verify_audit_trail_completeness()\n            }\n  -\n        def _aggregate_decisions_by_party(self) -> Dict[str, int]:\n            \"\"\"Calculate decision counts per accountable party\"\"\"\n            aggregation = {}\n            for decision in self.decision_audit_trail:\n                party = decision.accountable_party\n                aggregation[party] = aggregation.get(party, 0) + 1\n            return aggregation\n  -\n        def _calculate_redress_metrics(self) -> Dict:\n            \"\"\"Compute statistics on redress procedures\"\"\"\n            # Placeholder implementation\n            return {\"total_redress_cases\": 0, \"resolution_rate\": 0.0}\n  -\n        def _verify_audit_trail_completeness(self) -> float:\n            \"\"\"Check completeness of audit trail documentation\"\"\"\n            if not self.decision_audit_trail:\n                return 0.0\n            complete_records = sum(1 for d in self.decision_audit_trail\n                                  if d.accountable_party and d.audit_metadata)\n            return complete_records / len(self.decision_audit_trail)\n    ```\n    -\n  - ### Use Cases\n  - ### Use Case 1: Financial Services Credit Scoring\n    - **Scenario**: Bank deploys AI system for automated credit decisioning affecting loan approvals\n    - **Implementation**: Accountability framework assigns responsibility to Chief Risk Officer for model decisions, implements audit trail logging every credit decision with justification, enables applicants to contest decisions through formal redress procedure, requires quarterly algorithmic audits by independent third party\n    - **Benefits**: Regulatory compliance with fair lending laws, reduced discrimination risk, increased consumer trust, clear liability assignment, contestable decisions meeting due process requirements\n    - **Standards**: EU AI Act Article 14, IEEE P2863 Substantive Requirements, OECD Principle 1.3\n    -\n  - ### Technical Considerations\n  - ### Performance\n    - **Audit Trail Overhead**: Comprehensive decision logging increases storage requirements approximately 15-30% depending on metadata captured\n    - **Query Performance**: Indexed audit trails enable sub-second retrieval of decision provenance for redress procedures\n    - **Scalability**: Distributed accountability registries support horizontal scaling for high-volume decision systems\n    -\n  - ### Challenges and Solutions\n  - ### Challenge: Attribution in Complex AI Systems\n    - **Problem**: Multi-stage AI pipelines involving data providers, model developers, and deployment organisations create ambiguous responsibility assignment\n    - **Solution**: Implement accountability chain documentation mapping each system component to responsible party; use smart contracts to codify multi-party accountability agreements; establish joint liability frameworks for shared accountability scenarios\n    - **Standard Reference**: IEEE P2863 Section 5.3 - Procedural Governance Requirements\n    -\n  - ### Best Practices\n    1. **Comprehensive Responsibility Mapping**: Document all accountable parties across entire AI system lifecycle from data collection through deployment and monitoring\n    2. **Granular Audit Trails**: Log individual algorithmic decisions with sufficient metadata to enable complete reconstruction and explanation\n    3. **Accessible Redress Mechanisms**: Implement user-friendly contestation procedures requiring no technical expertise to invoke\n    4. **Independent Oversight**: Establish governance bodies with technical expertise and independence from AI system developers\n    5. **Regular Impact Assessments**: Conduct periodic algorithmic impact evaluations identifying emerging accountability gaps\n    6. **Proactive Transparency**: Publish accountability frameworks, responsible parties, and oversight procedures publicly\n    7. **Continuous Monitoring**: Implement automated compliance verification detecting accountability framework violations in real-time\n    8. **Stakeholder Engagement**: Involve affected communities in accountability framework design and oversight procedures\n    -\n  - ### Standards Alignment\n  - ### ISO/IEC Standards\n    - **ISO/IEC 42001:2023**: AI Management System requiring accountability as core governance component (Clause 5.3 - Leadership and Commitment)\n    - **ISO/IEC 23894:2023**: AI Risk Management emphasising accountability in risk treatment (Section 7.4 - Risk Treatment)\n    - **ISO/IEC TR 24028:2020**: AI Trustworthiness identifying accountability as key trustworthiness characteristic",
   "backlinks": [],
   "wiki_links": [
+    "AIEthicsDomain",
     "RegulatoryCompliance",
-    "EthicalFramework",
-    "IEEE P2863",
     "ConceptualLayer",
-    "AIEthicsDomain",
+    "IEEE P2863",
+    "EthicalFramework",
     "AIGovernancePrinciple"
   ],
   "ontology": {