Are we learning yet?
Metaheuristics
Metaheuristics aim to generate or select a heuristic for an optimization problem, particularly where the set of sample solutions is too large to completely sample.
A handful of metaheuristic crates exists in the Rust ecosystem, most notably several libraries that make it straight-forward to write genetic or evolutionary algorithms.
argmin [ crate · repo · docs ]
Mathematical optimization in pure Rust
oxigen [ crate · repo · docs ]
Fast, parallel, extensible and adaptable genetic algorithm library.
mop [ crate · repo · docs ]
Flexible and modular single or multi-objective solver for contiguous and discrete problems
simplers_optimization [ crate · repo · docs ]
A Rust implementation of the Simple(x) black-box optimization algorithm.
rsgenetic [ crate · repo · docs ]
A library providing genetic algorithm execution.
metaheuristics [ crate · repo · docs ]
Find approximate solutions to your optimisation problem using metaheuristics algorithms
darwin-rs [ crate · repo · docs ]
Evolutionary algorithms library written in Rust.
genetic [ crate · repo · docs ]
a little lib to use genetic algorithm
revonet [ crate · repo · docs ]
Rust implementation of real-coded genetic algorithm for solving optimization problems and training of neural networks. The latter is also known as neuroevolution.
evo [ crate · repo · docs ]
Evolutionary Algorithm Library for Rust
pengowen123/eant2 [ repo · docs ]
EANT2 is an metaheuristicsary algorithm that uses the Common Genetic Encoding (CGE) to encode neural networks and CMA-ES as a weight optimizer. EANT2 evolves the topology of the Neural Network and CMA-ES tries to find the optimal weights for each topology.
Looking for something you didn't find? Try asking on Zulip chat.