evobandits 0.0.5

Optimization algorithm combining genetic algorithms and multi-armed bandits for stochastic problems

EvoBandits is a cutting-edge optimization algorithm that merges genetic algorithms and multi-armed bandit strategies to efficiently solve stochastic problems.

---

EvoBandits (Evolutionary Multi-Armed Bandits) is an innovative optimization algorithm designed to tackle stochastic problems with high efficiency. By combining genetic algorithms with multi-armed bandit mechanisms, EvoBandits offers a unique, reinforcement learning-based approach to solving complex, large-scale optimization issues. Whether you're working in operations research, machine learning, or data science, EvoBandits provides a robust, scalable solution for optimizing your stochastic models.

Usage

To install EvoBandits, add the following to your Cargo.toml file:

pip install evobandits

from evobandits import EvoBandits

def test_function(number: list) -> float:
    # your function here

if __name__ == '__main__':
    bounds = [(-5, 10), (-5, 10)]
    evobandits = EvoBandits(test_function, bounds)
    evaluation_budget = 10000
    result = evobandits.optimize(evaluation_budget)
    print(result)

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you want to change.

License

tbd

Credit

Deniz Preil wrote the initial EvoBandits code, which Timo Kühne and Jonathan Laib rewrote.

Citing EvoBandits

If you use EvoBandits in your research, please cite the following paper:

Preil, D., & Krapp, M. (2024). Genetic Multi-Armed Bandits: A Reinforcement Learning Inspired Approach for Simulation Optimization. IEEE Transactions on Evolutionary Computation.