Emergent Quantized Communication
Preprint
Abstract
The field of emergent communication aims to understand the characteristics of communication as it emerges from artificial agents solving tasks that require information exchange. Communication with discrete messages is considered a desired characteristic, for both scientific and applied reasons. However, training a multi-agent system with discrete communication is not straightforward, requiring either reinforcement learning algorithms or relaxing the discreteness requirement via a continuous approximation such as the Gumbel-softmax. Both these solutions result in poor performance compared to fully continuous communication. In this work, we propose an alternative approach to achieve discrete communication -- quantization of communicated messages. Using message quantization allows us to train the model end-to-end, achieving superior performance in multiple setups. Moreover, quantization is a natural framework that runs the gamut from continuous to discrete communication. Thus, it sets the ground for a broader view of multi-agent communication in the deep learning era.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- A Survey of Quantization Methods for Efficient Neural Network Inference2022
- Few-Shot Intent Detection via Contrastive Pre-Training and Fine-Tuning2021
- Benchmarking Commercial Intent Detection Services with Practice-Driven Evaluations2021
- Capacity, Bandwidth, and Compositionality in Emergent Language Learning2020
- Multi-agent Communication meets Natural Language: Synergies between Functional and Structural Language Learning2020
- Efficient Intent Detection with Dual Sentence Encoders2020
- EGG: a toolkit for research on Emergence of lanGuage in Games2019
- Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning1992
- The Philosophy of PCMProceedings of the IRE, 1948