Manifold Grace

Research & Knowledge · AI and machine-learning development

AI categories, architectures and methods

Three separate registers. Each distinct architecture or method has one row; application domains and overarching concepts are kept apart. Historical project and paper initiatives remain linked without presenting proposals as completed research.

15 category and umbrella entries · 32 architecture and system-design entries · 14 method and algorithm entries. The original 40 base records are accounted for once each, with 6 additional research methods added.

Updated canonical taxonomy: The four-level register now separates learning paradigms from algorithms and architectures and tracks evidence status, aliases, and cross-references. Open the canonical four-level tables · Matching Excel-compatible CSV. The three legacy tables below remain for research continuity.

AI category types and domains

15 entries

Application areas, capabilities and umbrella families. These are not counted as individual model architectures. Project and paper links identify representative proposed initiatives.

#AI type / methodClassificationCategory or scopeWikipediaManifold Grace projectPaper initiativeResearch status
1Computer visionApplication domainImage and video understandingWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
2Natural language processingApplication domainLanguage interpretation and generationWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
3Predictive analyticsApplication domainPrediction from historical and contextual dataWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
4Expert and knowledge-based AIAI approach / system domainSymbolic knowledge and rule-guided reasoningWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
5Generative AICapability domainGeneration of text, images, audio and other outputsWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
6Multimodal AICapability domainJoint learning or reasoning across modalitiesWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
7Agentic AISystem domainTool use, planning and multi-step actionsWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
8Robotics and embodied AIApplication domainPerception, motion and action in physical settingsWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
9Graph and relational AITechnical domainRelationships, graph-structured data and inferenceWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
10Knowledge retrieval and question answeringApplication domainSource retrieval and answer generationWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
11AI-assisted decision and optimisationApplication domainRanking, inference and decision supportWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
12Speech and audio AIApplication domainSpeech recognition, audio perception and synthesisWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
13Neural-network-based AIUmbrella model familyA broad class containing many distinct architectures; not itself one unique architectureWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
14Classification and pattern recognitionApplication domainPredicting classes and identifying patternsWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project
15AI-assisted process automationApplication / integration domainWorkflows that optionally incorporate AI modelsWikipediaProjectPaperDomain / umbrella; project link is a representative example, not an exclusive project

Matching Excel-compatible CSV

Model and system architectures

32 entries

Distinct model classes and technical system designs already in the existing register. System integration patterns are labelled, not confused with neural model architecture.

#AI type / methodClassificationCategory or scopeWikipediaManifold Grace projectPaper initiativeResearch status
1CNN feature hierarchymodel architectureComputer visionWikipediaProjectPaperProposed project and paper; no independent validation claimed
2Vision Transformermodel architectureComputer visionWikipediaProjectPaperProposed project and paper; no independent validation claimed
3Object detection networkmodel architecture classComputer visionWikipediaProjectPaperProposed project and paper; no independent validation claimed
4Semantic segmentation networkmodel architecture classComputer visionWikipediaProjectPaperProposed project and paper; no independent validation claimed
5Instance segmentation networkmodel architecture classComputer visionWikipediaProjectPaperProposed project and paper; no independent validation claimed
6Encoder Transformermodel architectureLanguage and knowledgeWikipediaProjectPaperProposed project and paper; no independent validation claimed
7Autoregressive Transformermodel architectureLanguage and knowledgeWikipediaProjectPaperProposed project and paper; no independent validation claimed
8Retrieval-augmented generationsystem integration patternLanguage and knowledgeWikipediaProjectPaperProposed project and paper; no independent validation claimed
9Sequence-to-sequence Transformermodel architectureLanguage and knowledgeWikipediaProjectPaperProposed project and paper; no independent validation claimed
10Knowledge-graph question answeringapplication architectureLanguage and knowledgeWikipediaProjectPaperProposed project and paper; no independent validation claimed
11Temporal forecasting Transformermodel architecturePredictive and tabularWikipediaProjectPaperProposed project and paper; no independent validation claimed
12State-space sequence modelmodel architecturePredictive and tabularWikipediaProjectPaperProposed project and paper; no independent validation claimed
13Graph convolutional networkmodel architectureGraphs and relational dataWikipediaProjectPaperProposed project and paper; no independent validation claimed
14Graph attention networkmodel architectureGraphs and relational dataWikipediaProjectPaperProposed project and paper; no independent validation claimed
15Relational graph neural networkmodel architectureGraphs and relational dataWikipediaProjectPaperProposed project and paper; no independent validation claimed
16Temporal graph neural networkmodel architectureGraphs and relational dataWikipediaProjectPaperProposed project and paper; no independent validation claimed
17Neural relational database modelresearch model/system conceptGraphs and relational dataWikipediaProjectPaperProposed project and paper; no independent validation claimed
18Denoising diffusion modelgenerative model architectureGenerative and multimodalWikipediaProjectPaperProposed project and paper; no independent validation claimed
19Latent diffusion modelgenerative model architectureGenerative and multimodalWikipediaProjectPaperProposed project and paper; no independent validation claimed
20Generative adversarial networkgenerative model architectureGenerative and multimodalWikipediaProjectPaperProposed project and paper; no independent validation claimed
21Variational autoencodergenerative model architectureGenerative and multimodalWikipediaProjectPaperProposed project and paper; no independent validation claimed
22Vision-language modelmultimodal model architecture classGenerative and multimodalWikipediaProjectPaperProposed project and paper; no independent validation claimed
23Tool-using LLM agentagent system architectureAgents and automationWikipediaProjectPaperProposed project and paper; no independent validation claimed
24Plan-and-execute agentagent orchestration patternAgents and automationWikipediaProjectPaperProposed project and paper; no independent validation claimed
25Multi-agent orchestrationmulti-agent system patternAgents and automationWikipediaProjectPaperProposed project and paper; no independent validation claimed
26AI-enabled BPM workflowbusiness workflow architectureAgents and automationWikipediaProjectPaperProposed project and paper; no independent validation claimed
27Event-driven decision agentagent system architectureAgents and automationWikipediaProjectPaperProposed project and paper; no independent validation claimed
28Vision-based robotic policyrobotics policy architectureRobotics and controlWikipediaProjectPaperProposed project and paper; no independent validation claimed
29Spiking neural controllerneural control architectureRobotics and controlWikipediaProjectPaperProposed project and paper; no independent validation claimed
30Rule-based expert systemsymbolic AI systemLogic, search and hybrid decisionWikipediaProjectPaperProposed project and paper; no independent validation claimed
31Bayesian networkprobabilistic graphical modelLogic, search and hybrid decisionWikipediaProjectPaperProposed project and paper; no independent validation claimed
32Neuro-symbolic reasoning systemhybrid reasoning architectureLogic, search and hybrid decisionWikipediaProjectPaperProposed project and paper; no independent validation claimed

Matching Excel-compatible CSV

Statistical methods, learning paradigms and algorithms

14 entries

Methods and learning approaches, including survival analysis. A method may be used inside several architectures but is not itself an architecture just for that reason.

#AI type / methodClassificationCategory or scopeWikipediaManifold Grace projectPaper initiativeResearch status
1Gradient-boosted treesmachine-learning algorithm familyPredictive and tabularWikipediaProjectPaperProposed project and paper; no independent validation claimed
2Generalised additive modelstatistical model familyPredictive and tabularWikipediaProjectPaperProposed project and paper; no independent validation claimed
3Survival and hazard modelstatistical model familyPredictive and tabularWikipediaProjectPaperProposed project and paper; no independent validation claimed
4Model-predictive controllercontrol methodRobotics and controlWikipediaProjectPaperProposed project and paper; no independent validation claimed
5Reinforcement learning policylearning-policy methodRobotics and controlWikipediaProjectPaperProposed project and paper; no independent validation claimed
6Imitation learning policylearning-policy methodRobotics and controlWikipediaProjectPaperProposed project and paper; no independent validation claimed
7Constraint-satisfaction solverconstraint-solving algorithmLogic, search and hybrid decisionWikipediaProjectPaperProposed project and paper; no independent validation claimed
8Case-based reasoning systemreasoning methodLogic, search and hybrid decisionWikipediaProjectPaperProposed project and paper; no independent validation claimed
9Survival analysisStatistical time-to-event methodTime-to-event and censored observationsWikipediaProjectPaperProposed research initiative; not a unique model architecture
10Fuzzy logicLogical reasoning frameworkApproximate reasoning with degrees of truthWikipediaProjectPaperProposed research initiative; not a unique model architecture
11Supervised learningMachine-learning training paradigmLearning from labelled examplesWikipediaProjectPaperProposed research initiative; not a unique model architecture
12Unsupervised learningMachine-learning training paradigmFinding patterns without target labelsWikipediaProjectPaperProposed research initiative; not a unique model architecture
13Reinforcement learningMachine-learning training paradigmSequential decisions with reward feedbackWikipediaProjectPaperProposed research initiative; not a unique model architecture
14Deep learningMachine-learning approachLearning representations using multilayer modelsWikipediaProjectPaperProposed research initiative; not a unique model architecture

Matching Excel-compatible CSV

Research evidence and limits

The classification is an evidence-based working taxonomy, not a universal canonical count. A few system architectures can also be described as methods at a different level; the register assigns one primary home for counting and records that qualification. All proposed project and paper links remain working templates, not publications.

Survival analysis, censoring and Cox regression (2023) · Supervised algorithms · Unsupervised algorithms · NLP domain · AI/ML/deep learning distinctions.

Audit date: 14 November 2026. Original 1,000 proposed configurations preserved separately. Links to Wikipedia are introductory references, not independent proof of research validity.