Semantics and Spatiality of Emergent Communication
This work addresses the challenge of ensuring interpretable communication in AI agents, offering insights for researchers in emergent communication, though it is incremental in refining existing objectives.
The paper tackled the problem of opaque communication protocols in artificial agents by identifying semantic consistency as a prerequisite for meaningful communication, proving that reconstruction objectives avoid inconsistent protocols while discrimination can allow them, with experiments validating these theoretical results.
When artificial agents are jointly trained to perform collaborative tasks using a communication channel, they develop opaque goal-oriented communication protocols. Good task performance is often considered sufficient evidence that meaningful communication is taking place, but existing empirical results show that communication strategies induced by common objectives can be counterintuitive whilst solving the task nearly perfectly. In this work, we identify a goal-agnostic prerequisite to meaningful communication, which we term semantic consistency, based on the idea that messages should have similar meanings across instances. We provide a formal definition for this idea, and use it to compare the two most common objectives in the field of emergent communication: discrimination and reconstruction. We prove, under mild assumptions, that semantically inconsistent communication protocols can be optimal solutions to the discrimination task, but not to reconstruction. We further show that the reconstruction objective encourages a stricter property, spatial meaningfulness, which also accounts for the distance between messages. Experiments with emergent communication games validate our theoretical results. These findings demonstrate an inherent advantage of distance-based communication goals, and contextualize previous empirical discoveries.