HyperbolicNeuralNetwork<T>

Models & Types · AiDotNet.NeuralNetworks

Represents a Hyperbolic Neural Network for learning hierarchical representations in Poincare ball space.

Inheritance: NeuralNetworkBase<T> → DeclaredModelLayoutBase<T> → VectorModelLayoutBase<T> → HyperbolicNeuralNetwork<T>

For Beginners

Hyperbolic neural networks are designed for data that has a natural hierarchy or tree-like structure. Examples include:

  • Taxonomies (e.g., animal kingdom classification)
  • Organizational hierarchies
  • Social networks with community structures
  • Knowledge graphs

In hyperbolic space, the “distance” near the center is smaller than near the edges, allowing hierarchies to be represented more efficiently than in regular flat space. Points near the center represent “root” concepts, while points near the edge represent more specific “leaf” concepts.

How It Works

A Hyperbolic Neural Network operates in hyperbolic space (specifically the Poincare ball model) rather than Euclidean space. This allows it to naturally capture hierarchical and tree-like structures in data with lower distortion than traditional networks.

Example

var architecture = new NeuralNetworkArchitecture<float>(inputFeatures: 8, outputSize: 4);
var options = new HyperbolicNeuralNetworkOptions { Curvature = -1.0 };
var trainX = Tensor<float>.CreateRandom(4, 8);
var trainY = Tensor<float>.CreateRandom(4, 4);
var result = new AiModelBuilder<float, Tensor<float>, Tensor<float>>()
    .ConfigureModel(new HyperbolicNeuralNetwork<float>(architecture, options: options))
    .Build(trainX, trainY);
var input = Tensor<float>.CreateRandom(new[] { 1, 8 });
var output = result.Predict(input);

Constructors

Constructor Summary
HyperbolicNeuralNetwork() Initializes a new instance of the HyperbolicNeuralNetwork class.
HyperbolicNeuralNetwork(NeuralNetworkArchitecture<T> architecture, IGradientBasedOptimizer<T, Tensor<T>, Tensor<T>> optimizer, ILossFunction<T> lossFunction, HyperbolicNeuralNetworkOptions options)

Properties

Property Summary
bool SupportsTraining { get; } Serializes hyperbolic neural network-specific data to a binary writer.

Methods

Method Summary
ModelMetadata<T> GetModelMetadata() Retrieves metadata about the hyperbolic neural network model.
Tensor<T> Forward(Tensor<T> input) Performs a forward pass through the network with the given input tensor.