Towards Graph Representation Learning in Emergent Communication

Abstract

Recent findings in neuroscience suggest that the human brain represents information in a geometric structure (for instance, through conceptual spaces). In order to communicate, we flatten the complex representation of entities and their attributes into a single word or a sentence. In this paper we use graph convolutional networks to support the evolution of language and cooperation in multi-agent systems. Motivated by an image-based referential game, we propose a graph referential game with varying degrees of complexity, and we provide strong baseline models that exhibit desirable properties in terms of language emergence and cooperation. We show that the emerged communication protocol is robust, that the agents uncover the true factors of variation in the game, and that they learn to generalize beyond the samples encountered during training.

The first two authors contributed equally. Accepted at the Reinforcement Learning in Games workshop at AAAI 2020

References in corpus (8)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Inductive Representation Learning on Large Graphs
• Categorical Reparameterization with Gumbel-Softmax
• Learning to Communicate with Deep Multi-Agent Reinforcement Learning
• The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
• Emergence of Grounded Compositional Language in Multi-Agent Populations
• Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks
• Multi-Agent Cooperation and the Emergence of (Natural) Language

Cited by in corpus (2)

• Catalytic Role Of Noise And Necessity Of Inductive Biases In The Emergence Of Compositional Communication
• Structural Inductive Biases in Emergent Communication