Developed by Shahrokh Shahi
Georgia Institute of Technology
2018-2019
- Download the latest release on MATLAB File Exchange:
This tool includes an implementation of five gradient-based algorithms, including
- (1) Steepest Descend Algorithm (SDA),
- (2) Conjugate Gradient Algorithm (CGA),
- (3) Newton's Method,
- (4) Davidson-Fletcher-Powell (DFP) Method,
- (5) Broyden–Fletcher–Goldfarb–Shanno (BFGS) Method,
for multivariate function optimization. For two variable functions, a user-friendly interactive interface is available which portrays the optimization procedure step-by-step. This toolbox can also serve as an educational tool to understand mainstream optimization techniques.
- For more information about the details, see my blog post.
- An animation of the toolbox
- Run "OPTool.mltbx" in MATLAB and complete the installation
- Run OPTool in MATLAB command window
- Enjoy! (the app will be run with a pre-set values as an example)
Please note that the GUI is built and tested in Windows; there might be some visuall diferences in macOS (for instance, the test may appear smaller)