Pseudo-visibility: A Game Mechanic Involving Willful Ignorance

• Published in: Proceedings of the ... International Florida Artificial Intelligence Research Society Conference Vol. 35
• Main Authors: Alexander, Samuel, Pedersen, Arthur Paul
• Format: Journal Article Conference Proceeding
• Language: English
• Published: George A. Smathers Libraries 04.05.2022
• ISSN: 2334-0762; 2334-0754
• DOI: 10.32473/flairs.v35i.130652

We present a game mechanic called pseudo-visibility for games inhabited by non-player characters (NPCs) driven by reinforcement learning (RL). NPCs are incentivized to pretend they cannot see pseudo-visible players: the training environment simulates an NPC to determine how the NPC would act if the pseudo-visible player were invisible, and penalizes the NPC for acting differently. NPCs are thereby trained to selectively ignore pseudo-visible players, except when they judge that the reaction penalty is an acceptable tradeoff (e.g., a guard might accept the penalty in order to protect a treasure because losing the treasure would hurt even more). We describe an RL agent transformation which allows RL agents that would not otherwise do so to perform some limited self-reflection to learn the training environments in question.