A simple C++ library for multi-armed bandit simulations with multiple policy implementations.
Multi-armed bandit simulation library
Prioritizes clarity and ease of modification over optimization, as stated in the README, making it ideal for educational use and experimental prototyping.
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Requires only a modern C++ compiler and included tools like waf and cmdline.h, ensuring straightforward setup without library conflicts, per the README.
Implements key policies such as DMED, KL-UCB, Thompson sampling, and UCB variants, providing a broad range of bandit strategies for experimentation.
Includes a Python/Matplotlib-based plot tool for visualizing simulation results, adding convenience for analysis without additional setup.
Explicitly only supports Linux/GNU C++ environments and does not check for Windows or MacOS, restricting usability for cross-platform projects.
The README admits it's 'not optimized for speed,' making it inefficient for large-scale or real-time simulations compared to optimized libraries.
Beyond the basic README, there's no detailed API documentation or tutorials, which can hinder learning, extension, and troubleshooting.