Emergent Communication in a Multi-Modal, Multi-Step Referential Game

Abstract

Inspired by previous work on emergent communication in referential games, we propose a novel multi-modal, multi-step referential game, where the sender and receiver have access to distinct modalities of an object, and their information exchange is bidirectional and of arbitrary duration. The multi-modal multi-step setting allows agents to develop an internal communication significantly closer to natural language, in that they share a single set of messages, and that the length of the conversation may vary according to the difficulty of the task. We examine these properties empirically using a dataset consisting of images and textual descriptions of mammals, where the agents are tasked with identifying the correct object. Our experiments indicate that a robust and efficient communication protocol emerges, where gradual information exchange informs better predictions and higher communication bandwidth improves generalization.

arXiv commentPublished as a conference paper at ICLR 2018. 12 pages

References in corpus (10)

• Learning Cooperative Visual Dialog Agents with Deep Reinforcement Learning
• Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks
• Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols
• Reasoning About Pragmatics with Neural Listeners and Speakers
• Learning to Play Guess Who? and Inventing a Grounded Language as a Consequence
• End-to-end optimization of goal-driven and visually grounded dialogue systems
• Virtual Embodiment: A Scalable Long-Term Strategy for Artificial Intelligence Research
• A Paradigm for Situated and Goal-Driven Language Learning
• GuessWhat?! Visual object discovery through multi-modal dialogue
• Translating Neuralese

Cited by in corpus (15)

• Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense
• Maintaining cooperation in complex social dilemmas using deep reinforcement learning
• Emergent Translation in Multi-Agent Communication
• Emergence of Linguistic Communication from Referential Games with Symbolic and Pixel Input
• Consequentialist conditional cooperation in social dilemmas with imperfect information
• Graph Convolutional Networks for Classification with a Structured Label Space
• Emergence of Pragmatics from Referential Game between Theory of Mind Agents
• Emergence of Theory of Mind Collaboration in Multiagent Systems
• Emergent Language: A Survey and Taxonomy
• Interactive Reinforcement Learning for Object Grounding via Self-Talking
• A perspective on multi-agent communication for information fusion
• The emergence of visual semantics through communication games
• A Signal-Centric Perspective on the Evolution of Symbolic Communication
• On Memory Mechanism in Multi-Agent Reinforcement Learning
• Pow-Wow: A Dataset and Study on Collaborative Communication in Pommerman