Dynamic human-in-the-loop agent training

A system comprising:

• a controller including one or more processors configured to execute program instructions stored in a memory medium causing the one or more processors to:
• receive one or more tasks for a character to complete in one or more training sessions of an environment, where the character is controlled by a player, wherein the player is selectable between a human user and an agent implementing an active model by the controller;
• dynamically switch the player between the human user and the agent during the one or more training sessions, wherein at least one condition for switching the player is based on an introspection metric generated by the agent, where the introspection metric is associated with at least one of a confidence of the agent or observed actions when the player is the human user;
• capture observations from the character in the environment and corresponding actions of the character in the environment from the player during the one or more training sessions as experience data;
• and
• train one or more models for the agent based on the experience data, wherein the one or more models includes the active model.