Agent query techniques for multi-agent simulator platform

One or more computer-readable media having computer-executable instructions stored thereon for simulating a virtual world in a multi-agent simulator using an evolutionary framework, the instructions, when executed by one or more processors, causing a computing system to perform operations comprising:

• generating a virtual world comprising a set of agents, a first agent in the set of agents having:
• (1) one or more agent traits, and (2) one or more local interaction rules implemented by a first trained machine learning model;
• receiving, via a graphical user interface (GUI) of a computing device, experimenter input comprising a query and a set of input traits;
• instantiating the set of agents using the set of input traits;
• executing a first simulation session by causing the set of agents to generate a first output set;
• accessing an additional data source to generate world characteristics, the world characteristics comprising one or more of (1) an environmental condition and (2) a temporal condition;
• using the world characteristics to determine agent evolution logic;
• generating an evolved set of agents by evolving the set of agents according to the agent evolution logic by performing at least one of mutation of agent traits, selection of agents, crossover of agent characteristics, or deletion of agents that do not meet a fitness threshold;
• executing a second simulation session using an evolved set of agents to generate a second output set;
• and
• displaying at least a portion of the first output set and second output set at the GUI.

A computing system having one or more processors and one or more memory units having computer-executable instructions stored therein for simulating virtual environments in multi-agent simulator using an evolutionary framework, the instructions, when executed by the one or more processors, causing the computing system to:

• generate a virtual environment comprising a set of AI agents, a first agent in the set of AI agents having:
• (1) one or more agent traits, and (2) one or more local interaction rules implemented by a first trained AI model;
• receive experimenter input comprising a query and a set of input traits;
• instantiate the set of AI agents using the set of input traits;
• execute a first simulation session by causing the set of AI agents to generate a first output set;
• access an additional data source to generate world characteristics, the world characteristics comprising one or more of:
• (1) an environmental condition or (2) a temporal condition;
• use the world characteristics to determine agent evolution logic;
• generate an evolved set of AI agents by evolving the set of AI agents according to the agent evolution logic by performing at least one of:
• mutation of agent traits, selection of agents, crossover of agent characteristics, or deletion of agents that do not meet a fitness condition;
• execute a second simulation session using the evolved set of AI agents to generate a second output set;
• and
• transmit at least a portion of the first output set and second output set via an interface of the computing system.