Black-box Optimization Problems
- The black-box optimization problem in general refers to the optimization problem where the objective function (and/or constraints) does not have an analytical expression, and is lack of useable gradient information. The objective function values and the corresponding input variable values are the only information that can be utilized to search for an optimal solution.
Purpose of the RABBO benchmark
- The RABBO benchmark is maintained by the DAMO Academy's Decision Intelligence Laboratory. The leaderboard will be updated periodically and the benchmark problem set will be expanded to involve more real-world problems that can be solved by black-box optimization algorithms.
How to participate
- You can donwload the RABBO benchmark suite using "git clone", get used to the RABBO's APIs by running the examples, and start developing your own algorithms following the guidelines.
- To participate the official evaluation held by Tianchi, you need to first wrap up your algorithm by building a docker image and then submit it at the Tianchi's webpage.
- Welcome to join us and feel free to give us feedback and suggestions on improving RABBO.
