Audience
Game developers that want to use AI to dynamically generate NPC responses
About Chaos Box
An AI engine that uses the Chaos Box algorithm to analyze real-time player inputs and dynamically generate NPC responses and new storylines based on Deep Reinforcement Learning. It can support massively emergent behaviors for AIs and NPCs in games without even a single script.
Different from DeepMind and OpenAI, we are exploring more in the multi-agent and multi-objective scenario as for Reinforcement Learning. In this scenario, the behavior patterns for each different agent should coordinate and connect with each other to generate reasonable and dramatic story experience, helping the game production team significantly enhance every, no matter simple or complicated, statistics that can measure the user conversion rate in these experience-driven scenarios. Furthermore, based on the structure of game cloud service with client SDK, the game production teams can integrate our service during any stage of the game project with less effort, giving them a better user experience.