ADMMOptimizer<T, TInput, TOutput>

Models & Types · AiDotNet.Optimizers

Implements the Alternating Direction Method of Multipliers (ADMM) optimization algorithm.

Inheritance: OptimizerBase<T, TInput, TOutput> → GradientBasedOptimizerBase<T, TInput, TOutput> → ADMMOptimizer<T, TInput, TOutput>

For Beginners

ADMM is like solving a complex puzzle by breaking it into smaller, manageable pieces. It’s particularly good at handling problems with constraints or when you want to distribute the computation across multiple processors.

How It Works

ADMM is an algorithm for solving convex optimization problems, particularly useful for large-scale and distributed optimization. It combines the benefits of dual decomposition and augmented Lagrangian methods.

Example

using AiDotNet;
using AiDotNet.Data.Loaders;
using AiDotNet.Enums;
using AiDotNet.NeuralNetworks;
using AiDotNet.Optimizers;
using AiDotNet.Tensors.LinearAlgebra;

var rng = new Random(0);
var trainX = new Tensor<double>(new[] { 32, 8 });
var trainY = new Tensor<double>(new[] { 32, 2 });
for (int i = 0; i < 32; i++)
{
    for (int j = 0; j < 8; j++) trainX[new[] { i, j }] = rng.NextDouble();
    trainY[new[] { i, i % 2 }] = 1.0;
}

var model = new NeuralNetwork<double>(new NeuralNetworkArchitecture<double>(
    inputFeatures: 8, numClasses: 2, complexity: NetworkComplexity.Simple));

var result = await new AiModelBuilder<double, Tensor<double>, Tensor<double>>()
    .ConfigureModel(model)
    .ConfigureOptimizer(new ADMMOptimizer<double, Tensor<double>, Tensor<double>>(model))
    .ConfigureDataLoader(DataLoaders.FromTensors(trainX, trainY))
    .BuildAsync();

Console.WriteLine("Trained with ADMMOptimizer.");

Constructors

Constructor Summary
ADMMOptimizer(IFullModel<T, TInput, TOutput> model, ADMMOptimizerOptions<T, TInput, TOutput> options, IEngine engine) Initializes a new instance of the ADMMOptimizer class.

Methods

Method Summary
ADMMOptimizer<T, TInput, TOutput> CreateForFunction(ADMMOptimizerOptions<T, TInput, TOutput> options) Creates an ADMM optimizer for minimizing a plain function, with no model attached.
OptimizationAlgorithmOptions<T, TInput, TOutput> GetOptions() Retrieves the current options of the optimizer.
OptimizationResult<T, TInput, TOutput> Optimize(OptimizationInputData<T, TInput, TOutput> inputData) Performs the optimization process using the ADMM algorithm.
Vector<T> UpdateParameters(Vector<T> parameters, Vector<T> gradient) Applies one linearized-ADMM iteration to a flat parameter vector.
void UpdateParametersGpu(IGpuBuffer parameters, IGpuBuffer gradients, int parameterCount, IDirectGpuBackend backend) Updates parameters using GPU-accelerated ADMM.