CHON

as donto believed on

From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)

CHON has 139 facts recorded in Dontopedia across 15 references, with 10 live disagreements.

139 facts·106 predicates·15 sources·10 in dispute

Mostly:has feature(8), comprises component(6), rdf:type(5)

Full Name`fullName`

Clifford Harmonic Oscillator Network^[13]sourceall time · 499

Inbound mentions (35)

Other subjects in dontopedia point AT this entity as a value. These are inverse relationships — e.g. "X motherOf this subject" — and answer questions the forward facts can't. Grouped by predicate.

`intendsToBuild`Intends to Build(2)

Xenonfunex:xenonfun
Xenonfunex:xenonfun

`advocatesBuildingChon`Advocates Building Chon(1)

Xenonfunex:xenonfun

`announcedFeatureSet`Announced Feature Set(1)

Xenonfunex:xenonfun

`announcedFullFeatureSetShipped`Announced Full Feature Set Shipped(1)

Xenonfunex:xenonfun

`announcesCompletion`Announces Completion(1)

Xenonfunex:xenonfun

`assertsKnowledgeOfCHONCapabilities`Asserts Knowledge of Chon Capabilities(1)

Xenonfunex:xenonfun

`builtInto`Built Into(1)

Core Insight Paperex:core-insight-paper

`comparesMethod`Compares Method(1)

Comparison Tableex:comparison-table

`confirmsCapability`Confirms Capability(1)

4 Benchmarks Passingex:4-benchmarks-passing

`developed`Developed(1)

Xenonfunex:xenonfun

`discusses`Discusses(1)

Log Entry 2026 03 22 20 38ex:log-entry-2026-03-22-20-38

`evaluatedAsUndeniable`Evaluated As Undeniable(1)

Parameter Efficiencyex:parameter-efficiency

`evaluatesCHONAsReady`Evaluates Chon As Ready(1)

Xenonfunex:xenonfun

`hasBuiltInAnomalyDetection`Has Built in Anomaly Detection(1)

Chonex:chon

`hasConfigurableIntegrationSteps`Has Configurable Integration Steps(1)

Chonex:chon

`hasContinuousDepth`Has Continuous Depth(1)

Chonex:chon

`hasDAsFullMatrix`Has D As Full Matrix(1)

Chonex:chon

`hasGradeSeparatedReadout`Has Grade Separated Readout(1)

Chonex:chon

`includesMethod`Includes Method(1)

Comparison Tableex:comparison-table

`isBuilt`Is Built(1)

Chonex:chon

`isHead`Is Head(1)

Rul Headex:rul-head

`isNeuralODEStyle`Is Neural Ode Style(1)

Chonex:chon

`isPushed`Is Pushed(1)

Chonex:chon

`isReferencedAsBaseline`Is Referenced As Baseline(1)

Ugtex:ugt

`isRelatedProject`Is Related Project(1)

Microsoft Cliffordlayersex:microsoft-cliffordlayers

`isShipped`Is Shipped(1)

Chonex:chon

`isStackOfLayersWithInputOutputProjection`Is Stack of Layers With Input Output Projection(1)

Chonex:chon

`isTested`Is Tested(1)

Chonex:chon

`mapsEachGradeToPhysicalQuantity`Maps Each Grade to Physical Quantity(1)

Chonex:chon

`mentionsAdvantageOf`Mentions Advantage of(1)

Message 1ex:message-1

`partOfSystem`Part of System(1)

Chon Layerex:chon-layer

`relevantTo`Relevant to(1)

Paper Bivector Noncommutativityex:paper-bivector-noncommutativity

`sufficientForChon`Sufficient for Chon(1)

90 Percent Building Blocksex:90-percent-building-blocks

`testsModels`Tests Models(1)

No Regressions Step 6ex:no-regressions-step-6

Other facts (133)

The long tail: predicates that appear too rarely to warrant their own section. Filter or scroll to find a specific one. Each row links to its source.

133 facts

Predicate	Value	Ref
Has Feature	Rul Head	[6]
Has Feature	Anomaly Detector Ema Dynamic Threshold	[6]
Has Feature	Data Loaders Csv Binary NASA Dir	[6]
Has Feature	Multi Scale Ω G Groups	[6]
Has Feature	Softplus Gamma	[6]
Has Feature	Strang Splitting Rodrigues	[6]
Has Feature	Time Varying Ω T	[6]
Has Feature	Anomaly detection built-in	[11]
Comprises Component	Chonlayer	[1]
Comprises Component	Dissipation Anisotropic	[1]
Comprises Component	Dissipation Isotropic	[1]
Comprises Component	Extract Features Function	[1]
Comprises Component	Forward Sequence Function	[1]
Comprises Component	Ode Residual Function	[1]
Rdf:type	Model	[10]
Rdf:type	Software Architecture	[11]
Rdf:type	Model	[12]
Rdf:type	Model	[13]
Rdf:type	Software Project	[15]
Superior in Parameter Count to	Lstm Cnn	[7]
Superior in Parameter Count to	Lstm	[7]
Superior in Parameter Count to	Bilstm Pycaret	[7]
Parameter to Recover	Ω	[12]
Parameter to Recover	D	[12]
Parameter to Recover	K	[12]
Has Fewer Parameters Than	Lstm Autoencoder	[14]
Has Fewer Parameters Than	Lstm Cnn	[14]
Has Fewer Parameters Than	Bilstm Pycaret	[14]
Has Configurable Integration Steps	Chon	[1]
Has Configurable Integration Steps	true	[11]
Has Test Count	10	[1]
Has Test Count	10	[11]
Uses	Rodrigues Integrator	[3]
Uses	Softplus Bounded Gamma	[3]
Numerically Stable Due to	Softplus Bounded Γ	[3]
Numerically Stable Due to	Exact Rodrigues Integrator	[3]
Has Detection Lead Time	13 hours	[7]
Has Detection Lead Time	13 hours	[14]
Has Parameter Count	119	[7]
Has Parameter Count	119	[14]
Enables Anomaly Detection Via Residual	Ode Residual	[1]
Is Stack of Layers With Input Output Projection	Chon	[1]
Is Ready for	Spatiotemporal Tasks	[1]
Is Pushed	Chon	[1]
Is Neural Ode Style	Chon	[1]
Has Architecture Line Count Approx	500	[1]
Has Built in Anomaly Detection	Chon	[1]
Is Merging to	Master Branch	[1]
Has D As Full Matrix	Chon	[1]
Teleologically Designed for Spatiotemporal Tasks	Spatiotemporal Tasks	[1]
Supports Anisotropic Dissipation	Dissipation Anisotropic	[1]
Provides Advanced Dissipation Over	Ugt	[1]
Presupposes Existence of Cl30	Cl 3 0	[1]
Asserts Superiority in Anomaly Detection	Ugt	[1]
Is Built	Chon	[1]
Has Grade Separated Readout	Chon	[1]
Contrasts With Ugt in Dissipation	Ugt	[1]
Embodies Continuous Time Dynamics	Neural Ode	[1]
Enables Shear Dependent Learning	Dissipation Anisotropic	[1]
Implies Suitability for Physics Modeling	Physical Quantities	[1]
Has Continuous Depth	Chon	[1]
Differs From by Full Matrix D	Ugt	[1]
Maps Each Grade to Physical Quantity	Chon	[1]
Is Tested	Chon	[1]
Built Using Existing Blocks	90 Percent Building Blocks	[2]
Next Development Target	Architecture	[2]
Built by	You Recently	[3]
Can Recover	known dynamics	[3]
Physically Insightful	null	[3]
Contains Term	‑[Ω,Ψ]	[3]
Converges on	Ω Parameters	[3]
Numerically Stable	null	[3]
Uses Higher Lr for	Omega Parameter	[3]
Built Recently	null	[3]
Can Train on Right Now	Synthetic Dataset	[4]
Is Fully Functional	null	[4]
Has Loss	0.013	[4]
Achieves First Real World Result	NASA Ims Bearing Failure Detection	[4]
Detects Bearing Failure	null	[4]
Exists As Implemented Model	Existential Commitment	[5]
Is Learning	True	[5]
Has Recovered Omega0 With10 Error	Empirically	[5]
Is Shipped	Chon	[6]
Supports Multi Datasets	Data Loaders	[6]
Does Detection With Lead Time	True	[7]
Has Key Advantages	Key Advantages List	[7]
Avoids Need for	Feature Engineering	[7]
Teleologically Aims at	Anomaly Detection With Lead Time	[7]
Uses Zero Feature Engineering	True	[7]
Causes Interpretability of	Bearing Dynamics	[7]
Has Training Time	<1 second	[7]
Superior in Detection Lead Time Reporting to	Published Methods	[7]
Inputs Raw Rms Directly Into	Cl 3 0 Multivector	[7]
Requires Zero Feature Engineering	True	[7]
Provides Detection Lead Time	13 hours	[7]
Preferred Over	Deep Learning Baselines	[7]
Is Physically Interpretable	True	[7]
Has Accuracy F1 Method	Clean threshold crossing	[7]
Has1000x Fewer Parameters Than	Lstm Baselines	[7]
Contrasts With	Published Methods	[7]

Timeline axis is valid_time — when each source says the fact was true in the world, not when Dontopedia learned about it. Retracted rows are kept for provenance; coloured stripes indicate the context kind.

139 undated

—

enablesAnomalyDetectionViaResidualblah/watt-activation/part-498

ex:ode-residual

—

hasConfigurableIntegrationStepsblah/watt-activation/part-498

ex:chon

—

isStackOfLayersWithInputOutputProjectionblah/watt-activation/part-498

ex:chon

—

isReadyForblah/watt-activation/part-498

ex:spatiotemporal-tasks

—

isPushedblah/watt-activation/part-498

ex:chon

—

isNeuralODEStyleblah/watt-activation/part-498

ex:chon

—

hasArchitectureLineCountApproxblah/watt-activation/part-498

500

—

hasBuiltInAnomalyDetectionblah/watt-activation/part-498

ex:chon

—

isMergingToblah/watt-activation/part-498

ex:master-branch

—

hasDAsFullMatrixblah/watt-activation/part-498

ex:chon

—

teleologicallyDesignedForSpatiotemporalTasksblah/watt-activation/part-498

ex:spatiotemporal-tasks

—

supportsAnisotropicDissipationblah/watt-activation/part-498

ex:dissipation-anisotropic

—

providesAdvancedDissipationOverblah/watt-activation/part-498

ex:ugt

—

presupposesExistenceOfCl30blah/watt-activation/part-498

ex:cl-3-0

—

hasTestCountblah/watt-activation/part-498

—

comprisesComponentblah/watt-activation/part-498

ex:chonlayer

—

comprisesComponentblah/watt-activation/part-498

ex:dissipation-anisotropic

—

comprisesComponentblah/watt-activation/part-498

ex:dissipation-isotropic

—

comprisesComponentblah/watt-activation/part-498

ex:extract-features-function

—

comprisesComponentblah/watt-activation/part-498

ex:forward-sequence-function

—

comprisesComponentblah/watt-activation/part-498

ex:ode-residual-function

—

assertsSuperiorityInAnomalyDetectionblah/watt-activation/part-498

ex:ugt

—

isBuiltblah/watt-activation/part-498

ex:chon

—

hasGradeSeparatedReadoutblah/watt-activation/part-498

ex:chon

—

contrastsWithUGTInDissipationblah/watt-activation/part-498

ex:ugt

—

embodiesContinuousTimeDynamicsblah/watt-activation/part-498

ex:neural-ode

—

enablesShearDependentLearningblah/watt-activation/part-498

ex:dissipation-anisotropic

—

impliesSuitabilityForPhysicsModelingblah/watt-activation/part-498

ex:physical-quantities

—

hasContinuousDepthblah/watt-activation/part-498

ex:chon

—

differsFromByFullMatrixDblah/watt-activation/part-498

ex:ugt

—

mapsEachGradeToPhysicalQuantityblah/watt-activation/part-498

ex:chon

—

isTestedblah/watt-activation/part-498

ex:chon

—

builtUsingExistingBlocksblah/watt-activation/part-497

ex:90-percent-building-blocks

—

nextDevelopmentTargetblah/watt-activation/part-497

ex:architecture

—

builtByblah/watt-activation/part-502

ex:you-recently

—

usesblah/watt-activation/part-502

ex:rodrigues-integrator

—

canRecoverblah/watt-activation/part-502

known dynamics

—

physicallyInsightfulblah/watt-activation/part-502

null

—

numericallyStableDueToblah/watt-activation/part-502

ex:softplus-bounded-γ

—

containsTermblah/watt-activation/part-502

‑[Ω,Ψ]

—

numericallyStableDueToblah/watt-activation/part-502

ex:exact-rodrigues-integrator

—

convergesOnblah/watt-activation/part-502

ex:Ω-parameters

—

numericallyStableblah/watt-activation/part-502

null

—

usesHigherLrForblah/watt-activation/part-502

ex:omega-parameter

—

usesblah/watt-activation/part-502

ex:softplus-bounded-gamma

—

builtRecentlyblah/watt-activation/part-502

null

—

canTrainOnRightNowblah/watt-activation/part-503

ex:synthetic-dataset

—

isFullyFunctionalblah/watt-activation/part-503

null

—

hasLossblah/watt-activation/part-503

0.013

—

achievesFirstRealWorldResultblah/watt-activation/part-503

ex:nasa-ims-bearing-failure-detection

—

detectsBearingFailureblah/watt-activation/part-503

null

—

existsAsImplementedModelblah/watt-activation/part-501

ex:existential-commitment

—

isLearningblah/watt-activation/part-501

ex:true

—

hasRecoveredOmega0With10Errorblah/watt-activation/part-501

ex:empirically

—