Class: Inference

Bayesian inference of model parameters from an observation, via MCMC. A standalone, first-class concept (NOT an Optimization): it produces a POSTERIOR, not a point estimate, using priors + a likelihood + a sampler instead of a loss + optimizer. It runs the SAME differentiable forward model, wrapped in a probabilistic model (sample priors -> forward -> likelihood).

URI: tvbo:Inference

 classDiagram
    class Inference
    click Inference href "./Inference.html"
      Inference : depends_on
        
          
    
        
        
        Inference --> "0..1" Algorithm : depends_on
        click Algorithm href "./Algorithm.html"
    

        
      Inference : description
        
          
    
        
        
        Inference --> "0..1" String : description
        click String href "../http://www.w3.org/2001/XMLSchema#string.html"
    

        
      Inference : execution
        
          
    
        
        
        Inference --> "0..1" ExecutionConfig : execution
        click ExecutionConfig href "./ExecutionConfig.html"
    

        
      Inference : integration
        
          
    
        
        
        Inference --> "0..1" Integrator : integration
        click Integrator href "./Integrator.html"
    

        
      Inference : label
        
          
    
        
        
        Inference --> "0..1" String : label
        click String href "../http://www.w3.org/2001/XMLSchema#string.html"
    

        
      Inference : likelihood
        
          
    
        
        
        Inference --> "1" Likelihood : likelihood
        click Likelihood href "./Likelihood.html"
    

        
      Inference : name
        
          
    
        
        
        Inference --> "1" String : name
        click String href "../http://www.w3.org/2001/XMLSchema#string.html"
    

        
      Inference : num_chains
        
          
    
        
        
        Inference --> "0..1" Integer : num_chains
        click Integer href "../http://www.w3.org/2001/XMLSchema#integer.html"
    

        
      Inference : num_samples
        
          
    
        
        
        Inference --> "0..1" Integer : num_samples
        click Integer href "../http://www.w3.org/2001/XMLSchema#integer.html"
    

        
      Inference : num_warmup
        
          
    
        
        
        Inference --> "0..1" Integer : num_warmup
        click Integer href "../http://www.w3.org/2001/XMLSchema#integer.html"
    

        
      Inference : priors
        
          
    
        
        
        Inference --> "*" Prior : priors
        click Prior href "./Prior.html"
    

        
      Inference : sampler
        
          
    
        
        
        Inference --> "0..1" String : sampler
        click String href "../http://www.w3.org/2001/XMLSchema#string.html"
    

        
      Inference : seed
        
          
    
        
        
        Inference --> "0..1" Integer : seed
        click Integer href "../http://www.w3.org/2001/XMLSchema#integer.html"

Class Properties

Property Value
Class URI tvbo:Inference

Slots

Name Cardinality and Range Description Inheritance
name 1
xsd:string		 Globally unique identifier for the entity direct
label 0..1
xsd:string		 direct
description 0..1
xsd:string		 direct
priors *
Prior		 Priors over the inferred parameters, keyed by each parameter’s dotted name (e direct
likelihood 1
Likelihood		 Observation model p(data sim(theta))
sampler 0..1
xsd:string		 MCMC sampler (‘nuts’) direct
num_samples 0..1
xsd:integer		 Posterior samples to draw direct
num_warmup 0..1
xsd:integer		 Warmup/adaptation steps before sampling direct
num_chains 0..1
xsd:integer		 Independent MCMC chains direct
seed 0..1
xsd:integer		 PRNG seed for reproducible sampling direct
execution 0..1
ExecutionConfig		 Per-inference execution config (overrides experiment defaults) direct
integration 0..1
Integrator		 Forward-model integration settings (overrides experiment defaults) direct
depends_on 0..1
Algorithm		 Optional algorithm whose result state seeds the inference direct

Usages

used by used in type used
SimulationExperiment inferences range Inference

Identifier and Mapping Information

Schema Source

  • from schema: https://w3id.org/tvbo

Mappings

Mapping Type Mapped Value
self tvbo:Inference
native tvbo:Inference

LinkML Source

Direct

name: Inference
description: 'Bayesian inference of model parameters from an observation, via MCMC.
  A standalone, first-class concept (NOT an Optimization): it produces a POSTERIOR,
  not a point estimate, using priors + a likelihood + a sampler instead of a loss
  + optimizer. It runs the SAME differentiable forward model, wrapped in a probabilistic
  model (sample priors -> forward -> likelihood).'
from_schema: https://w3id.org/tvbo
slots:
- name
- label
- description
attributes:
  priors:
    name: priors
    description: Priors over the inferred parameters, keyed by each parameter's dotted
      name (e.g. "Generic2dOscillator.eta" or "PulseInput.amplitude"). The KEYS are
      the inferred parameters — there is no separate free-parameter list.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    domain_of:
    - Inference
    range: Prior
    multivalued: true
    inlined: true
  likelihood:
    name: likelihood
    description: Observation model p(data | sim(theta)).
    from_schema: https://w3id.org/tvbo
    rank: 1000
    domain_of:
    - Inference
    range: Likelihood
    required: true
    inlined: true
  sampler:
    name: sampler
    description: MCMC sampler ('nuts'). Each backend maps its own way.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: string(nuts)
    domain_of:
    - Inference
    range: string
  num_samples:
    name: num_samples
    description: Posterior samples to draw.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: integer(1000)
    domain_of:
    - Inference
    range: integer
  num_warmup:
    name: num_warmup
    description: Warmup/adaptation steps before sampling.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: integer(1000)
    domain_of:
    - Inference
    range: integer
  num_chains:
    name: num_chains
    description: Independent MCMC chains.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: integer(1)
    domain_of:
    - Inference
    range: integer
  seed:
    name: seed
    description: PRNG seed for reproducible sampling.
    from_schema: https://w3id.org/tvbo
    ifabsent: integer(0)
    domain_of:
    - GraphGenerator
    - Distribution
    - Noise
    - Inference
    range: integer
  execution:
    name: execution
    description: Per-inference execution config (overrides experiment defaults).
    from_schema: https://w3id.org/tvbo
    domain_of:
    - Optimization
    - Exploration
    - Inference
    - Algorithm
    - Continuation
    - SimulationExperiment
    range: ExecutionConfig
    inlined: true
  integration:
    name: integration
    description: Forward-model integration settings (overrides experiment defaults).
    from_schema: https://w3id.org/tvbo
    domain_of:
    - Optimization
    - Inference
    - SimulationExperiment
    range: Integrator
    inlined: true
  depends_on:
    name: depends_on
    description: Optional algorithm whose result state seeds the inference.
    from_schema: https://w3id.org/tvbo
    domain_of:
    - Optimization
    - Inference
    - Algorithm
    range: Algorithm
    inlined: false
class_uri: tvbo:Inference

Induced

name: Inference
description: 'Bayesian inference of model parameters from an observation, via MCMC.
  A standalone, first-class concept (NOT an Optimization): it produces a POSTERIOR,
  not a point estimate, using priors + a likelihood + a sampler instead of a loss
  + optimizer. It runs the SAME differentiable forward model, wrapped in a probabilistic
  model (sample priors -> forward -> likelihood).'
from_schema: https://w3id.org/tvbo
attributes:
  priors:
    name: priors
    description: Priors over the inferred parameters, keyed by each parameter's dotted
      name (e.g. "Generic2dOscillator.eta" or "PulseInput.amplitude"). The KEYS are
      the inferred parameters — there is no separate free-parameter list.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    owner: Inference
    domain_of:
    - Inference
    range: Prior
    multivalued: true
    inlined: true
  likelihood:
    name: likelihood
    description: Observation model p(data | sim(theta)).
    from_schema: https://w3id.org/tvbo
    rank: 1000
    owner: Inference
    domain_of:
    - Inference
    range: Likelihood
    required: true
    inlined: true
  sampler:
    name: sampler
    description: MCMC sampler ('nuts'). Each backend maps its own way.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: string(nuts)
    owner: Inference
    domain_of:
    - Inference
    range: string
  num_samples:
    name: num_samples
    description: Posterior samples to draw.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: integer(1000)
    owner: Inference
    domain_of:
    - Inference
    range: integer
  num_warmup:
    name: num_warmup
    description: Warmup/adaptation steps before sampling.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: integer(1000)
    owner: Inference
    domain_of:
    - Inference
    range: integer
  num_chains:
    name: num_chains
    description: Independent MCMC chains.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: integer(1)
    owner: Inference
    domain_of:
    - Inference
    range: integer
  seed:
    name: seed
    description: PRNG seed for reproducible sampling.
    from_schema: https://w3id.org/tvbo
    ifabsent: integer(0)
    owner: Inference
    domain_of:
    - GraphGenerator
    - Distribution
    - Noise
    - Inference
    range: integer
  execution:
    name: execution
    description: Per-inference execution config (overrides experiment defaults).
    from_schema: https://w3id.org/tvbo
    owner: Inference
    domain_of:
    - Optimization
    - Exploration
    - Inference
    - Algorithm
    - Continuation
    - SimulationExperiment
    range: ExecutionConfig
    inlined: true
  integration:
    name: integration
    description: Forward-model integration settings (overrides experiment defaults).
    from_schema: https://w3id.org/tvbo
    owner: Inference
    domain_of:
    - Optimization
    - Inference
    - SimulationExperiment
    range: Integrator
    inlined: true
  depends_on:
    name: depends_on
    description: Optional algorithm whose result state seeds the inference.
    from_schema: https://w3id.org/tvbo
    owner: Inference
    domain_of:
    - Optimization
    - Inference
    - Algorithm
    range: Algorithm
    inlined: false
  name:
    name: name
    description: Globally unique identifier for the entity.
    from_schema: https://w3id.org/tvbo/common
    slot_uri: schema:name
    identifier: true
    owner: Inference
    domain_of:
    - BrainAtlas
    - CommonCoordinateSpace
    - ParcellationEntity
    - DBSProtocol
    - ClinicalScale
    - ClinicalScore
    - SoftwarePackage
    - SoftwareRequirement
    - SoftwareEnvironment
    - Event
    - Tractogram
    - MeasureSpec
    - NamedArray
    - GraphGenerator
    - Binding
    - ProcedureStep
    - File
    - StateValue
    - Observation
    - Dynamics
    - StateVariable
    - Distribution
    - Parameter
    - CouplingInput
    - Argument
    - Function
    - FunctionCall
    - Callable
    - DerivedParameter
    - DerivedVariable
    - DataSource
    - OptimizationStage
    - Exploration
    - Inference
    - Prior
    - Likelihood
    - UpdateRule
    - Algorithm
    - Option
    - BranchSwitch
    - Continuation
    - Coupling
    range: string
    required: true
  label:
    name: label
    from_schema: https://w3id.org/tvbo
    rank: 1000
    slot_uri: rdfs:label
    owner: Inference
    domain_of:
    - ParcellationTerminology
    - Subject
    - Session
    - Dataset
    - Contact
    - SoftwareEnvironment
    - Equation
    - Stimulus
    - Event
    - Parcellation
    - Tractogram
    - Matrix
    - Phenotype
    - Network
    - Node
    - Edge
    - Observation
    - Dynamics
    - StateVariable
    - Parameter
    - Function
    - FunctionCall
    - DerivedVariable
    - RandomStream
    - DataSource
    - OptimizationStage
    - Exploration
    - ExplorationAxis
    - FreeParameter
    - Inference
    - AlgorithmStage
    - TuningObjective
    - Continuation
    - Coupling
    - RegionMapping
    - SimulationExperiment
    - SimulationStudy
    - TimeSeries
    - NDArray
    - SpatialDomain
    - Mesh
    - SpatialField
    - FieldStateVariable
    - DifferentialOperator
    - BoundaryCondition
    - PDESolver
    - PDE
    range: string
  description:
    name: description
    from_schema: https://w3id.org/tvbo
    rank: 1000
    slot_uri: dcterms:description
    owner: Inference
    domain_of:
    - Dataset
    - ClinicalScore
    - SoftwarePackage
    - SoftwareRequirement
    - SoftwareEnvironment
    - Equation
    - Stimulus
    - Event
    - Tractogram
    - Matrix
    - Phenotype
    - MeasureSpec
    - NamedArray
    - Network
    - GraphGenerator
    - ProcedureStep
    - File
    - Node
    - Edge
    - Observation
    - Dynamics
    - StateVariable
    - Parameter
    - CouplingInput
    - Argument
    - Function
    - FunctionCall
    - Callable
    - DerivedParameter
    - DerivedVariable
    - RandomStream
    - DataSource
    - OptimizationStage
    - Exploration
    - ExplorationAxis
    - FreeParameter
    - Inference
    - Prior
    - Likelihood
    - UpdateRule
    - AlgorithmStage
    - TuningObjective
    - Algorithm
    - BranchSwitch
    - Continuation
    - Integrator
    - Coupling
    - RegionMapping
    - SimulationExperiment
    - Study
    - TimeSeries
    - NDArray
    - SpatialDomain
    - Mesh
    - SpatialField
    - FieldStateVariable
    - BoundaryCondition
    - PDESolver
    - PDE
    range: string
class_uri: tvbo:Inference