Log In Sign Up

Multitasking Inhibits Semantic Drift

by   Athul Paul Jacob, et al.

When intelligent agents communicate to accomplish shared goals, how do these goals shape the agents' language? We study the dynamics of learning in latent language policies (LLPs), in which instructor agents generate natural-language subgoal descriptions and executor agents map these descriptions to low-level actions. LLPs can solve challenging long-horizon reinforcement learning problems and provide a rich model for studying task-oriented language use. But previous work has found that LLP training is prone to semantic drift (use of messages in ways inconsistent with their original natural language meanings). Here, we demonstrate theoretically and empirically that multitask training is an effective counter to this problem: we prove that multitask training eliminates semantic drift in a well-studied family of signaling games, and show that multitask training of neural LLPs in a complex strategy game reduces drift and while improving sample efficiency.


Countering Language Drift with Seeded Iterated Learning

Supervised learning methods excel at capturing statistical properties of...

Supervised Seeded Iterated Learning for Interactive Language Learning

Language drift has been one of the major obstacles to train language mod...

Countering Language Drift via Visual Grounding

Emergent multi-agent communication protocols are very different from nat...

Skill Induction and Planning with Latent Language

We present a framework for learning hierarchical policies from demonstra...

How to Do Things without Words: Modeling Semantic Drift of Emoji

Emoji have become a significant part of our informal textual communicati...

Learning with Latent Language

The named concepts and compositional operators present in natural langua...

Cooperative image captioning

When describing images with natural language, the descriptions can be ma...