Class: Algorithm

A complete specification of an iterative parameter tuning algorithm. Combines update rules, objectives, observations, and hyperparameters.

URI: tvbo:class/Algorithm

 classDiagram
    class Algorithm
    click Algorithm href "./Algorithm.html"
      Algorithm : apply_every
        
          
    
        
        
        Algorithm --> "0..1" Integer : apply_every
        click Integer href "../http://www.w3.org/2001/XMLSchema#integer.html"
    

        
      Algorithm : depends_on
        
          
    
        
        
        Algorithm --> "*" Algorithm : depends_on
        click Algorithm href "./Algorithm.html"
    

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

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

        
      Algorithm : functions
        
          
    
        
        
        Algorithm --> "*" FunctionCall : functions
        click FunctionCall href "./FunctionCall.html"
    

        
      Algorithm : hyperparameters
        
          
    
        
        
        Algorithm --> "*" Parameter : hyperparameters
        click Parameter href "./Parameter.html"
    

        
      Algorithm : includes
        
          
    
        
        
        Algorithm --> "*" AlgorithmInclude : includes
        click AlgorithmInclude href "./AlgorithmInclude.html"
    

        
      Algorithm : learning_rate
        
          
    
        
        
        Algorithm --> "0..1" Float : learning_rate
        click Float href "../http://www.w3.org/2001/XMLSchema#float.html"
    

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

        
      Algorithm : learning_rate_warmup
        
          
    
        
        
        Algorithm --> "0..1" Boolean : learning_rate_warmup
        click Boolean href "../http://www.w3.org/2001/XMLSchema#boolean.html"
    

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

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

        
      Algorithm : objective
        
          
    
        
        
        Algorithm --> "0..1" TuningObjective : objective
        click TuningObjective href "./TuningObjective.html"
    

        
      Algorithm : observations
        
          
    
        
        
        Algorithm --> "*" Observation : observations
        click Observation href "./Observation.html"
    

        
      Algorithm : simulation_period
        
          
    
        
        
        Algorithm --> "0..1" Float : simulation_period
        click Float href "../http://www.w3.org/2001/XMLSchema#float.html"
    

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

        
      Algorithm : update_rules
        
          
    
        
        
        Algorithm --> "*" UpdateRule : update_rules
        click UpdateRule href "./UpdateRule.html"
    

        
      

Slots

Name	Cardinality and Range	Description	Inheritance
name	1 xsd:string		direct
description	0..1 xsd:string		direct
execution	0..1 ExecutionConfig	Per-algorithm execution configuration (overrides experiment-level defaults)	direct
type	0..1 xsd:string	Algorithm type: ‘fic’, ‘eib’, ‘homeostatic’, ‘custom’	direct
includes	* AlgorithmInclude	Include update rules from other algorithms with optional argument overrides	direct
objective	0..1 TuningObjective	What the algorithm optimizes for	direct
observations	* Observation	References to observations defined in the observations section	direct
update_rules	* UpdateRule	How parameters are updated each iteration	direct
hyperparameters	* Parameter	Additional algorithm-specific parameters	direct
learning_rate	0..1 xsd:float	Learning rate (eta) for the tuning algorithm	direct
learning_rate_warmup	0..1 xsd:boolean	Linear warmup of learning rate from 0 to learning_rate over n_iterations	direct
n_iterations	0..1 xsd:integer	Number of iterations to run	direct
learning_rate_schedule	0..1 xsd:string	Learning rate schedule: ‘constant’, ‘linear’, ‘exponential’	direct
simulation_period	0..1 xsd:float	Duration of each simulation step (e	direct
apply_every	0..1 xsd:integer	Apply update every N iterations	direct
functions	* FunctionCall	Function calls for tracking progress, computing metrics, etc	direct
depends_on	* Algorithm	Other algorithms that must run first (e	direct

Usages

used by	used in	type	used
Optimization	depends_on	range	Algorithm
AlgorithmInclude	algorithm	range	Algorithm
Algorithm	depends_on	range	Algorithm
SimulationExperiment	algorithms	range	Algorithm

Identifier and Mapping Information

Schema Source

• from schema: https://w3id.org/tvbo

Mappings

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

LinkML Source

Direct

name: Algorithm
description: A complete specification of an iterative parameter tuning algorithm.
  Combines update rules, objectives, observations, and hyperparameters.
from_schema: https://w3id.org/tvbo
slots:
- name
- description
attributes:
  execution:
    name: execution
    description: Per-algorithm execution configuration (overrides experiment-level
      defaults). Useful for setting random_seed per algorithm to ensure reproducibility.
    from_schema: https://w3id.org/tvbo
    domain_of:
    - Optimization
    - Exploration
    - Algorithm
    - SimulationExperiment
    range: ExecutionConfig
    inlined: true
  type:
    name: type
    description: 'Algorithm type: ''fic'', ''eib'', ''homeostatic'', ''custom'''
    from_schema: https://w3id.org/tvbo
    domain_of:
    - File
    - Aggregation
    - TuningObjective
    - Algorithm
    range: string
  includes:
    name: includes
    description: 'Include update rules from other algorithms with optional argument
      overrides. Unlike depends_on (sequential), includes means combined execution.
      Example: includes: [{algorithm: fic, arguments: [{name: eta, value: 0.1}]}]'
    from_schema: https://w3id.org/tvbo
    rank: 1000
    domain_of:
    - Algorithm
    range: AlgorithmInclude
    multivalued: true
    inlined: true
  objective:
    name: objective
    description: What the algorithm optimizes for
    from_schema: https://w3id.org/tvbo
    rank: 1000
    domain_of:
    - Algorithm
    range: TuningObjective
    inlined: true
  observations:
    name: observations
    description: References to observations defined in the observations section. Includes
      both simulated observations and external data (via data_source).
    from_schema: https://w3id.org/tvbo
    rank: 1000
    domain_of:
    - Algorithm
    - SimulationExperiment
    range: Observation
    multivalued: true
    inlined: false
  update_rules:
    name: update_rules
    description: How parameters are updated each iteration. When using 'includes',
      update_rules are inherited from included algorithms.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    domain_of:
    - Algorithm
    range: UpdateRule
    required: false
    multivalued: true
    inlined: true
  hyperparameters:
    name: hyperparameters
    description: Additional algorithm-specific parameters
    from_schema: https://w3id.org/tvbo
    domain_of:
    - OptimizationStage
    - Algorithm
    range: Parameter
    multivalued: true
    inlined: true
  learning_rate:
    name: learning_rate
    description: Learning rate (eta) for the tuning algorithm
    from_schema: https://w3id.org/tvbo
    domain_of:
    - OptimizationStage
    - Algorithm
    range: float
  learning_rate_warmup:
    name: learning_rate_warmup
    description: Linear warmup of learning rate from 0 to learning_rate over n_iterations.
      eta_effective = eta * (i+1) / n_iterations
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: boolean(false)
    domain_of:
    - Algorithm
    range: boolean
  n_iterations:
    name: n_iterations
    description: Number of iterations to run
    from_schema: https://w3id.org/tvbo
    rank: 1000
    domain_of:
    - Algorithm
    range: integer
  learning_rate_schedule:
    name: learning_rate_schedule
    description: 'Learning rate schedule: ''constant'', ''linear'', ''exponential'''
    from_schema: https://w3id.org/tvbo
    rank: 1000
    domain_of:
    - Algorithm
    range: string
  simulation_period:
    name: simulation_period
    description: Duration of each simulation step (e.g., one BOLD TR)
    from_schema: https://w3id.org/tvbo
    rank: 1000
    domain_of:
    - Algorithm
    range: float
  apply_every:
    name: apply_every
    description: Apply update every N iterations
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: integer(1)
    domain_of:
    - Algorithm
    range: integer
  functions:
    name: functions
    description: Function calls for tracking progress, computing metrics, etc. Each
      FunctionCall references a function from the experiment's functions section and
      specifies arguments for that specific algorithm context.
    from_schema: https://w3id.org/tvbo
    domain_of:
    - Dynamics
    - Algorithm
    - SimulationExperiment
    - PDE
    range: FunctionCall
    multivalued: true
    inlined: true
    inlined_as_list: true
  depends_on:
    name: depends_on
    description: Other algorithms that must run first (e.g., EIB depends on FIC)
    from_schema: https://w3id.org/tvbo
    domain_of:
    - Optimization
    - Algorithm
    range: Algorithm
    multivalued: true

Induced

name: Algorithm
description: A complete specification of an iterative parameter tuning algorithm.
  Combines update rules, objectives, observations, and hyperparameters.
from_schema: https://w3id.org/tvbo
attributes:
  execution:
    name: execution
    description: Per-algorithm execution configuration (overrides experiment-level
      defaults). Useful for setting random_seed per algorithm to ensure reproducibility.
    from_schema: https://w3id.org/tvbo
    alias: execution
    owner: Algorithm
    domain_of:
    - Optimization
    - Exploration
    - Algorithm
    - SimulationExperiment
    range: ExecutionConfig
    inlined: true
  type:
    name: type
    description: 'Algorithm type: ''fic'', ''eib'', ''homeostatic'', ''custom'''
    from_schema: https://w3id.org/tvbo
    alias: type
    owner: Algorithm
    domain_of:
    - File
    - Aggregation
    - TuningObjective
    - Algorithm
    range: string
  includes:
    name: includes
    description: 'Include update rules from other algorithms with optional argument
      overrides. Unlike depends_on (sequential), includes means combined execution.
      Example: includes: [{algorithm: fic, arguments: [{name: eta, value: 0.1}]}]'
    from_schema: https://w3id.org/tvbo
    rank: 1000
    alias: includes
    owner: Algorithm
    domain_of:
    - Algorithm
    range: AlgorithmInclude
    multivalued: true
    inlined: true
  objective:
    name: objective
    description: What the algorithm optimizes for
    from_schema: https://w3id.org/tvbo
    rank: 1000
    alias: objective
    owner: Algorithm
    domain_of:
    - Algorithm
    range: TuningObjective
    inlined: true
  observations:
    name: observations
    description: References to observations defined in the observations section. Includes
      both simulated observations and external data (via data_source).
    from_schema: https://w3id.org/tvbo
    rank: 1000
    alias: observations
    owner: Algorithm
    domain_of:
    - Algorithm
    - SimulationExperiment
    range: Observation
    multivalued: true
    inlined: false
  update_rules:
    name: update_rules
    description: How parameters are updated each iteration. When using 'includes',
      update_rules are inherited from included algorithms.
    from_schema: https://w3id.org/tvbo
    rank: 1000
    alias: update_rules
    owner: Algorithm
    domain_of:
    - Algorithm
    range: UpdateRule
    required: false
    multivalued: true
    inlined: true
  hyperparameters:
    name: hyperparameters
    description: Additional algorithm-specific parameters
    from_schema: https://w3id.org/tvbo
    alias: hyperparameters
    owner: Algorithm
    domain_of:
    - OptimizationStage
    - Algorithm
    range: Parameter
    multivalued: true
    inlined: true
  learning_rate:
    name: learning_rate
    description: Learning rate (eta) for the tuning algorithm
    from_schema: https://w3id.org/tvbo
    alias: learning_rate
    owner: Algorithm
    domain_of:
    - OptimizationStage
    - Algorithm
    range: float
  learning_rate_warmup:
    name: learning_rate_warmup
    description: Linear warmup of learning rate from 0 to learning_rate over n_iterations.
      eta_effective = eta * (i+1) / n_iterations
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: boolean(false)
    alias: learning_rate_warmup
    owner: Algorithm
    domain_of:
    - Algorithm
    range: boolean
  n_iterations:
    name: n_iterations
    description: Number of iterations to run
    from_schema: https://w3id.org/tvbo
    rank: 1000
    alias: n_iterations
    owner: Algorithm
    domain_of:
    - Algorithm
    range: integer
  learning_rate_schedule:
    name: learning_rate_schedule
    description: 'Learning rate schedule: ''constant'', ''linear'', ''exponential'''
    from_schema: https://w3id.org/tvbo
    rank: 1000
    alias: learning_rate_schedule
    owner: Algorithm
    domain_of:
    - Algorithm
    range: string
  simulation_period:
    name: simulation_period
    description: Duration of each simulation step (e.g., one BOLD TR)
    from_schema: https://w3id.org/tvbo
    rank: 1000
    alias: simulation_period
    owner: Algorithm
    domain_of:
    - Algorithm
    range: float
  apply_every:
    name: apply_every
    description: Apply update every N iterations
    from_schema: https://w3id.org/tvbo
    rank: 1000
    ifabsent: integer(1)
    alias: apply_every
    owner: Algorithm
    domain_of:
    - Algorithm
    range: integer
  functions:
    name: functions
    description: Function calls for tracking progress, computing metrics, etc. Each
      FunctionCall references a function from the experiment's functions section and
      specifies arguments for that specific algorithm context.
    from_schema: https://w3id.org/tvbo
    alias: functions
    owner: Algorithm
    domain_of:
    - Dynamics
    - Algorithm
    - SimulationExperiment
    - PDE
    range: FunctionCall
    multivalued: true
    inlined: true
    inlined_as_list: true
  depends_on:
    name: depends_on
    description: Other algorithms that must run first (e.g., EIB depends on FIC)
    from_schema: https://w3id.org/tvbo
    alias: depends_on
    owner: Algorithm
    domain_of:
    - Optimization
    - Algorithm
    range: Algorithm
    multivalued: true
  name:
    name: name
    from_schema: https://w3id.org/tvbo
    rank: 1000
    key: true
    alias: name
    owner: Algorithm
    domain_of:
    - BrainAtlas
    - CommonCoordinateSpace
    - ParcellationEntity
    - DBSProtocol
    - ClinicalScale
    - ClinicalScore
    - Tractogram
    - File
    - Observation
    - Dynamics
    - StateVariable
    - Distribution
    - Parameter
    - CouplingInput
    - Argument
    - Function
    - Callable
    - DerivedParameter
    - DerivedVariable
    - DataSource
    - OptimizationStage
    - Exploration
    - UpdateRule
    - Algorithm
    - Coupling
    - SoftwareEnvironment
    - SoftwareRequirement
    - SoftwarePackage
    range: string
    required: true
  description:
    name: description
    from_schema: https://w3id.org/tvbo
    rank: 1000
    alias: description
    owner: Algorithm
    domain_of:
    - ClinicalScore
    - Equation
    - Stimulus
    - Tractogram
    - Matrix
    - Network
    - File
    - Node
    - Edge
    - Observation
    - Dynamics
    - StateVariable
    - Parameter
    - CouplingInput
    - Argument
    - Function
    - Callable
    - DerivedParameter
    - DerivedVariable
    - RandomStream
    - DataSource
    - OptimizationStage
    - Exploration
    - UpdateRule
    - TuningObjective
    - Algorithm
    - Integrator
    - Coupling
    - RegionMapping
    - SimulationExperiment
    - SimulationStudy
    - TimeSeries
    - SoftwareEnvironment
    - SoftwareRequirement
    - SoftwarePackage
    - NDArray
    - SpatialDomain
    - Mesh
    - SpatialField
    - FieldStateVariable
    - BoundaryCondition
    - PDESolver
    - PDE
    range: string