A lightweight stochastic optimizer based on slime mold (Slime Mold Algorithm)
Slime
This is a Swift implementation of a Slime Mold Algorithm – a stochastic optimizer – generally based on this paper
The only dependency required by Slime is SwiftNumerics
Visual Examples
Searching for the global maxima of
-abs(x + 100000) - abs(y + 100000) + sin(10 * x)
Searching for the shortest path visiting 50 locations (traveling salesman)
Use the Slime
In a SwiftPM project:
Add the following line to the dependencies in your Package.swift file:
.package(url: "https://github.com/ejjonny/slime", from: "1.0.0"),
Add Slime as a dependency for your target:
.target(
name: "MyTarget",
dependencies: [
.product(name: "Slime", package: "Slime"),
]
),
Add import Slime to your swift file.
var slime = Slime(
populationSize: 10,
maxIterations: 100,
lowerBound: [-1, -1],
upperBound: [1, 1],
method: .minimize, // Use .maximize if higher fitness values are better
fitnessEvaluation: { vector in
let x = vector[0]
let y = vector[1]
// Return a fitness value Double using the proposed vector
}
)
slime.run() // This runs the fitness evaluation many times, among other busy work, & will usually be expensive
slime.bestCells // An array of the top 3 Cells. Use Cell.position for the associated vectors
This example is using a 2 dimensional solution space. The algorithm will work with any number of vector components if you’re looking for a solution in hyperspace.
More Info
“Slime mold algorithm (SMA) is a population-based optimization technique which is proposed based on the oscillation style of slime mold in nature. The SMA has a unique mathematical model that simulates positive and negative feedbacks of the propagation wave of slime mold. It has a dynamic structure with a stable balance between global and local search drifts.”
TODO:
- Readme walkthrough of the math used in the algorithm
- Explore some deterministic changes