Unsupervised Translation of Emergent Communication
This work addresses the challenge of understanding emergent communication for researchers in AI and linguistics, but it is incremental as it adapts existing UNMT techniques to a new domain.
The study tackled the problem of interpreting emergent communication (EC) by applying unsupervised neural machine translation (UNMT) to EC formed in referential games, finding that task complexity and semantic diversity affect translatability, with UNMT successfully translating EC without parallel data.
Emergent Communication (EC) provides a unique window into the language systems that emerge autonomously when agents are trained to jointly achieve shared goals. However, it is difficult to interpret EC and evaluate its relationship with natural languages (NL). This study employs unsupervised neural machine translation (UNMT) techniques to decipher ECs formed during referential games with varying task complexities, influenced by the semantic diversity of the environment. Our findings demonstrate UNMT's potential to translate EC, illustrating that task complexity characterized by semantic diversity enhances EC translatability, while higher task complexity with constrained semantic variability exhibits pragmatic EC, which, although challenging to interpret, remains suitable for translation. This research marks the first attempt, to our knowledge, to translate EC without the aid of parallel data.