Class AdditiveNoiseSimulation

java.lang.Object

edu.cmu.tetrad.sem.AdditiveNoiseSimulation

public class AdditiveNoiseSimulation
extends Object

AdditiveNoiseSimulation

Generates data from an additive-noise structural causal model (ANM):

X_j = f_j(Pa(X_j)) + N_j, with independent noise terms N_j.

Each f_j is represented by a randomly initialized MLP (parents-only input). Root nodes are generated as pure noise.

Parent values are passed through the activation function (bounding them) before entering each node's MLP, so f_j is a bounded-input function by construction. No sample-dependent rescaling is performed.

Constructor Summary

Constructors

Constructor	Description
AdditiveNoiseSimulation(Graph graph, int numSamples, Sampler sampler, int[] hiddenDimensions, double inputScale, Function<Double,Double> activationFunction)	Constructs a new AdditiveNoiseSimulation instance with the specified parameters.

Method Summary

Modifier and Type	Method	Description
DataSet	generateData()	Generates a synthetic dataset by simulating data propagation through a graph with additive noise.

Methods inherited from class Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

AdditiveNoiseSimulation

public AdditiveNoiseSimulation(Graph graph,
 int numSamples,
 Sampler sampler,
 int[] hiddenDimensions,
 double inputScale,
 Function<Double,Double> activationFunction)

Constructs a new AdditiveNoiseSimulation instance with the specified parameters.

Parameters:

graph - The causal graph representing the structural relationships.

numSamples - The number of data samples to generate.

sampler - The sampler for additive noise.

hiddenDimensions - The dimensions of hidden layers in the MLP.

inputScale - The scaling factor for input data.

activationFunction - The activation function for the MLP.

Method Details

generateData

public DataSet generateData()

Generates a synthetic dataset by simulating data propagation through a graph with additive noise. The method creates data for each node in the graph based on its topological order, parent relationships, and random multilayer perceptron (MLP) evaluations, along with additive noise.

Note: the returned dataset's columns are in topological order, not graph.getNodes() order.

Returns:

The generated synthetic dataset.