pdips / consensusbasedx.jl

A Julia package for consensus-based optimisation

ConsensusBasedX.jl: Consensus-Based Optimisation in Julia

ConsensusBasedX.jl is a gradient-free stochastic optimisation package for Julia, born out of Consensus.jl and CBXpy. It uses Consensus-Based Optimisation (CBO), a flavour of Particle Swarm Optimisation (PSO) first introduced by R. Pinnau, C. Totzeck, O. Tse, and S. Martin (2017). This is a method of global optimisation particularly suited for rough functions, where gradient descent would fail. It is useful for optimisation in higher dimensions. It also implements Consensus-Based Sampling (CBS), as introduced in J. A. Carrillo, F. Hoffmann, A. M. Stuart, and U. Vaes (2022).

How to install and use

To install ConsensusBasedX.jl, simply run

using Pkg; Pkg.add("ConsensusBasedX")

in the Julia REPL. You can then load the package in a script or in the REPL by running

using ConsensusBasedX

Basic minimisation

The main functionality of ConsensusBasedX.jl is function minimisation via CBO. It assumes you have defined a function f(x::AbstractVector) that takes a single vector argumemt x of length D = length(x).

For instance, if D = 2, you can minimise f by running:

minimise(f, D = 2)

Your full code might look like this:

using ConsensusBasedX
f(x) = x[1]^2 + x[2]^2
x = minimise(f, D = 2)

Basic sampling

ConsensusBasedX.jl also provides CBS. The package exports sample, which has the same syntax as minimise.

For instance, if D = 2, you can sample exp(-αf) by running:

out = sample(f, D = 2, extended_output=true)
out.sample

For more detailed explanations and full-code examples, see the documentation.

Bug reports, feature requests, and contributions

See the contribution guidelines.

Copyright © 2024 Dr Rafael Bailo and Purpose-Driven Interacting Particle Systems Group. MIT License.

consensusbasedx.jl's Issues

JOSS review comments

Hi there, and congrats on the package and JOSS submission!

I'm one of the reviewers and I'll put my remarks and questions below.

Register v1.0.0

@JuliaRegistrator register()

Register v1.0.0

@JuliaRegistrator register()

Register v1.1.0

@JuliaRegistrator register()

Register v1.0.0

@JuliaRegistrator register()

TagBot trigger issue

This issue is used to trigger TagBot; feel free to unsubscribe.

If you haven't already, you should update your TagBot.yml to include issue comment triggers.
Please see this post on Discourse for instructions and more details.

If you'd like for me to do this for you, comment TagBot fix on this issue.
I'll open a PR within a few hours, please be patient!

Register v1.2.0

@JuliaRegistrator register()

Add examples

It would be nice to add worked examples in the repository, illustrating the full pipeline: define initial particles, run CBO, plot the results. I'd like to work on implementing CBS, but am a little bit stuck getting an example for CBO to run fully. I really like the DefaultKeywordArguments technique, but it does make debugging more difficult.

The Default package is not publicly available

Hi Rafa,

I think this is the package you are developing, right?

Best wishes,

Urbain