Theme
Development of a general-purpose multi-agent deep reinforcement learning library
We aim to develop a versatile library that can be applied to various industries as “team-based AI.” Specifically, we will provide several cutting-edge models of “multi-agent deep reinforcement learning” that can be easily used by people in any industry. In particular, we plan to provide transformer-based models (the basis of ChatGPT is also a transformer), which have been attracting attention in recent years.
Following this, we will develop a demo to demonstrate the effectiveness of the library and show that “team-based AI” can be a powerful option for any industry. Additionally, we aim to enable robots to coordinate with each other using only text instructions by combining the library with large-scale language models.
To STARTUP LEAGUE WEB SITE
