Intention Inference Under Execution Noise: Separating Aleatoric and Epistemic Uncertainty in Social Dilemmas
2026-08-03 • Computer Science and Game Theory
Computer Science and Game TheoryMachine Learning
AI summaryⓘ
The authors study situations where actions in social games can be mistaken because of noise, meaning a bad action might be an accident rather than a true intention to harm. They use a special model (POMDP) that tries to guess what the other player really intends, rather than just what action was observed. Their method helps players decide when to cooperate or retaliate, but too much noise can still cause cooperation to break down. They show this is important particularly in games where understanding the other person's intention affects decisions.
noisy social dilemmasMarkov Decision Process (MDP)Partially Observable MDP (POMDP)active inference (AIF)Iterated Prisoner's Dilemmaintention inferenceepistemic costpragmatic costcooperation collapselatent states
Authors
Kival Mahadew, Jonathan Shock
Abstract
In noisy social dilemmas, intended actions are stochastically corrupted before execution, so an observed defection may reflect hostile intent or action error. Standard Markov Decision Process (MDP) formulations treat executed actions as states, structurally precluding this distinction and causing systematic over-retaliation. We introduce a Partially Observable MDP (POMDP) formulation encoding opponent intentions as latent states and executed actions as noisy observations, solved within the active inference (AIF) framework with a cost function that decomposes into epistemic and pragmatic components that jointly address inferring current intent and learning how intent evolves. In the Iterated Prisoner's Dilemma with symmetric noise, we derive a critical noise threshold governing cooperation collapse, connecting it to a fixed-point condition on learned priors. Experiments reveal that the value of intention inference is context-dependent: the POMDP provides consistent advantages against conditionally cooperative opponents, but mutual intention inference under sufficient noise produces correlated belief-driven collapse. The advantage is specific to games where intent attribution is decision-relevant.