Towards Multi-Agent Communication-Based Language Learning
Angeliki Lazaridou Nghia The Pham Marco Baroni University of Trento
Abstract
We propose an interactive multimodal framework for language learning. Instead of being passively exposed to large amounts of natural text, our learners (implemented as feed-forward neural networks) engage in cooperative referential games starting from a tabula rasa setup, and thus develop their own language from the need to communicate in order to succeed at the game. Preliminary experiments provide promising results, but also suggest that it is important to ensure that agents trained in this way do not develop an ad-hoc communication code only effective for the game they are playing.
中文速览
让机器自己"玩着学语言"——这项研究让两个神经网络智能体从零开始，通过玩指代游戏（referential game）自发地摸索出一套沟通符号：其中一个智能体要用一个词描述某个目标物体，另一个智能体则要凭这个词在一堆物体里认出它，两者只有配合成功才能得分。实验结果证明，这种多智能体互动框架确实可行，智能体能学会高效的指代策略；但关键隐患也随之暴露——它们自发发展出来的"暗语"只在彼此之间好使，与人类自然语言的语义对不上号。这说明，若想让这类智能体将来真正跟人类对话，就必须想办法把它们的通信协议"锚定"到自然语言的语义上，而这正是后续研究需要突破的核心挑战。
原文 arXiv:1605.07133;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1605.07133v1