Global Gradient Optimization Methods

This repository contains python implementation of algorithms for approach of finding global minimum of multidimensional functions.

Summary

The proposed method explores a potential of applying global optimization along the line employing Moore-Skelboe algorithm for finding global minimum on whole set. Proposed results compered to those achieved by vanilla Golden-section optimization.

Testing:

We tested our approach on the following functions:

• Rastrigin function
• Ackley function

Our model provides different optimization algorithms:

• Coordinate descent
• Gradient descent

References

[Interval Global Optimization] (https://www.researchgate.net/scientific-contributions/Helmut-Ratschek-70626150)

License

MIT License