Hello, my name is Zhou Shihai. We are an independent game team from China. I designed a deep reinforcement learning reward mechanism that mimics real human emotions. In effect, the agent seems to have an astrolabe. Players can combine the astrolabe (actually, data charts, etc.) to analyze the agent’s behavioral motives and emotional changes, thereby increasing the interaction between the player and the agent (although the astrolabe analysis of fate is not scientifically based, but The ideas inside can be used for reference).
The core gameplay is that through data visualization interaction, players can influence the action strategy adopted by the agent by using props or changing their behavior.
The game content includes two major pieces.
The first piece of content is gameplay and plot, 3D character action adventure, 2D background (there is actually not much new in this piece of content).
The second piece of content is the AI module, based on deep learning, including AI driver code, AI code emotional data and the visual interaction of the player, and deep learning visual component drag programming three game content.
Currently, our program team only has one program that uses spare time to produce the first piece of content. I would like to invite you to join our team responsible for the second piece of content.
The above content comes from Google Translate, there may be grammatical problems, please forgive me.