BinarySpikingActivation<T>

Models & Types · AiDotNet.ActivationFunctions

Implements the Binary Spiking activation function for neural networks, particularly for spiking neural networks.

Inheritance: ActivationFunctionBase<T> → BinarySpikingActivation<T>

For Beginners

Binary Spiking activation models how real neurons in your brain generate electrical pulses.

Unlike standard activation functions that output continuous values, Binary Spiking:

  • Outputs only 1 (spike/fire) or 0 (no spike)
  • Neurons “fire” only when their input exceeds a threshold
  • After firing, neurons typically have a “refractory period” before they can fire again

This activation creates the discrete, all-or-nothing behavior of biological neurons:

  • Input below threshold ? Output = 0 (neuron remains silent)
  • Input at or above threshold ? Output = 1 (neuron fires a spike)

Common uses include:

  • Spiking Neural Networks (SNNs)
  • Neuromorphic computing systems
  • Models that process temporal information
  • Energy-efficient neural networks for specialized hardware

Example

using AiDotNet.ActivationFunctions;

var activation = new BinarySpikingActivation<double>();
double y = activation.Activate(0.5);
Console.WriteLine($"BinarySpikingActivation: f(0.5) = {y:F4}");

Constructors

Constructor Summary
BinarySpikingActivation() Initializes a new instance of the Binary Spiking activation function with default parameters.
BinarySpikingActivation(T threshold, T derivativeSlope, T derivativeWidth) Initializes a new instance of the Binary Spiking activation function with custom parameters.

Methods

Method Summary
BinarySpikingActivation<T> WithThreshold(T newThreshold) Creates a new instance of the Binary Spiking activation with a different threshold.
Matrix<T> Derivative(Vector<T> input) Calculates the derivative of the Binary Spiking function for a tensor input.
T Activate(T x) Applies the Binary Spiking activation function to a tensor input.
T Derivative(T x) Calculates the derivative of the Binary Spiking function for a tensor input.
T GetThreshold() Gets the firing threshold value used by this activation function.
Tensor<T> Activate(Tensor<T> input) Applies the Binary Spiking activation function to a tensor input.
Tensor<T> Derivative(Tensor<T> input) Calculates the derivative of the Binary Spiking function for a tensor input.
Vector<T> Activate(Vector<T> input) Applies the Binary Spiking activation function to a tensor input.