Emergent Communication: New Frontiers


5th Workshop on Emergent Communication

News

April 27 Our ICLR Workshop Page is live with links to the Zoom and RocketChat

April 18 The Schedule is finalized!

April 7 Best Reviewer Award goes to Dylan Cope and Kevin Denamganaï

April 4 Best Paper Award goes to Situated Communication: A Solution to Over-communication between Artificial Agents

runner up Learning To Ground Decentralized Multi-Agent Communication with Contrastive Learning

runner up Emergent Communication Fine-tuning (EC-FT) for Pretrained Language Models

March 27 Accepted papers and reviews now available on OpenReview. Congratulations to all accepted authors!

Abstract

How is communication learned? How can communication be leveraged to improve AI systems? How does language evolve with interaction?

This one-day, online workshop will bring together researchers across disciplines (machine learning, philosophy, biology, linguistics, …) for discussions and talks. This year’s theme, New Frontiers, will explore novel uses and methods of communication: how can we apply methods from language emergence and multi-agent communication to other problems and fields? how can methods from other fields be applied here? The workshop aims to tackle grand challenges and foster interdisciplinary collaborations.

New This Year:

  • No poster sessions, all accepted authors will lead small discussion groups on their contributions

  • 1:1 mentoring session between senior researchers and student authors (and participants if there is availability)

  • Best paper award is a work of art by Prof. Simon Kirby (one of the invited speakers)

  • We are accepting previously published research in fields outside of ML

Invited Speakers

ICREA / UPF

Uni. of Edinburgh

UC Berkeley / Google

Conlanger (Language Inventor)

Stephen F. Austin State University

Abstract

When it comes to language, ML has for a long time only approached the problem of language understanding, and was striving to do so using supervised learning to find statistical regularities on large amounts of data. This narrow focus was leaving aside language emergence (within a population of autonomous agents, using reinforcement learning (RL), without relying on pre-recorded natural language data) and language grounding (in some other modalities, e.g. sight, with more complex inputs than text).

Thus, such approaches were failing to capture the functional and interactive aspects of both natural and artificial languages. Traditional ML approaches were ignoring the how and why of language use: to communicate and facilitate cooperation between autonomous agents. In contrast, research in Emergent Communication (EC) studies learning to communicate by interacting with other agents to solve collaborative tasks in complex and diverse environments.

Thanks to deep RL’s ability to handle complex data, EC is no longer confined to being studied by linguistics and theoretical AI practitioners, it can now be studied practically in complex multi-agent scenarios. Thus, a recent resurgence and standardization effort manifested itself in the form of four previous successful workshops (2017-2020). Those workshops have gathered the community to discuss how, when, to what end, and how fast communication emerges, producing research later published at top ML venues (e.g., ICLR, ICML, AAAI).

Following those breakthroughs, research directions rapidly branched out to create New Frontiers where applications in diverse domains are leveraging EC. We therefore believe that EC has significant potential to impact a wide range of disciplines both within AI (e.g. MARL, visual-question answering, explainability, robotics) and beyond (e.g. social linguistics, cognitive science, philosophy of language).

Organisers

Meta AI/UPF

Uni. of Cambridge

Niko Grupen

Cornell University

Mathieu Rita

INRIA

Uni. of York

Scientific Committee

Oxford University

Dalhousie University

HuggingFace

Sponsors