Enum: ExperimentalDesignMethodType

Computational and statistical methods used for experimental design in autonomous laboratory workflows

URI: valuesets:ExperimentalDesignMethodType

Permissible Values

Value	Description	Approach	Approaches
BAYESIAN_OPTIMIZATION	Probabilistic model-based optimization using Bayesian inference to guide experimental selection	surrogate model with acquisition function
ACTIVE_LEARNING	Machine learning approach that iteratively selects the most informative experiments to perform	query-based learning from unlabeled data
REINFORCEMENT_LEARNING	Machine learning approach where an agent learns optimal experimental strategies through trial and reward	reward-based sequential decision making
EVOLUTIONARY_ALGORITHM	Optimization method inspired by biological evolution using mutation, crossover, and selection	population-based metaheuristic
RANDOM_SEARCH	Experimental design using random sampling of the parameter space	random sampling
GRID_SEARCH	Systematic exploration of parameter space using a regular grid of experimental conditions	exhaustive grid evaluation
LATIN_HYPERCUBE_SAMPLING	Statistical sampling method that ensures even coverage of each parameter dimension	stratified random sampling
DESIGN_OF_EXPERIMENTS	Classical statistical methods for planning experiments including factorial and response surface designs		full factorial, fractional factorial, Box-Behnken, central composite
MULTI_OBJECTIVE_OPTIMIZATION	Optimization approach that simultaneously considers multiple competing objectives	Pareto-optimal solution search

Slots

Name	Description
experimental_design_method	Method used for experimental design in autonomous laboratories

Identifier and Mapping Information

Schema Source

from schema: https://w3id.org/valuesets

LinkML Source

name: ExperimentalDesignMethodType
instantiates:
- valuesets_meta:ValueSetEnumDefinition
description: Computational and statistical methods used for experimental design in
  autonomous laboratory workflows
title: Experimental Design Method Type
from_schema: https://w3id.org/valuesets
contributors:
- orcid:0000-0002-6601-2165
- https://github.com/anthropics/claude-code
status: DRAFT
rank: 1000
permissible_values:
  BAYESIAN_OPTIMIZATION:
    text: BAYESIAN_OPTIMIZATION
    description: Probabilistic model-based optimization using Bayesian inference to
      guide experimental selection
    annotations:
      approach:
        tag: approach
        value: surrogate model with acquisition function
  ACTIVE_LEARNING:
    text: ACTIVE_LEARNING
    description: Machine learning approach that iteratively selects the most informative
      experiments to perform
    annotations:
      approach:
        tag: approach
        value: query-based learning from unlabeled data
  REINFORCEMENT_LEARNING:
    text: REINFORCEMENT_LEARNING
    description: Machine learning approach where an agent learns optimal experimental
      strategies through trial and reward
    annotations:
      approach:
        tag: approach
        value: reward-based sequential decision making
  EVOLUTIONARY_ALGORITHM:
    text: EVOLUTIONARY_ALGORITHM
    description: Optimization method inspired by biological evolution using mutation,
      crossover, and selection
    annotations:
      approach:
        tag: approach
        value: population-based metaheuristic
  RANDOM_SEARCH:
    text: RANDOM_SEARCH
    description: Experimental design using random sampling of the parameter space
    annotations:
      approach:
        tag: approach
        value: random sampling
  GRID_SEARCH:
    text: GRID_SEARCH
    description: Systematic exploration of parameter space using a regular grid of
      experimental conditions
    annotations:
      approach:
        tag: approach
        value: exhaustive grid evaluation
  LATIN_HYPERCUBE_SAMPLING:
    text: LATIN_HYPERCUBE_SAMPLING
    description: Statistical sampling method that ensures even coverage of each parameter
      dimension
    annotations:
      approach:
        tag: approach
        value: stratified random sampling
  DESIGN_OF_EXPERIMENTS:
    text: DESIGN_OF_EXPERIMENTS
    description: Classical statistical methods for planning experiments including
      factorial and response surface designs
    annotations:
      approaches:
        tag: approaches
        value: full factorial, fractional factorial, Box-Behnken, central composite
  MULTI_OBJECTIVE_OPTIMIZATION:
    text: MULTI_OBJECTIVE_OPTIMIZATION
    description: Optimization approach that simultaneously considers multiple competing
      objectives
    annotations:
      approach:
        tag: approach
        value: Pareto-optimal solution search