Call for Papers

Some sample questions that we would like to explore are:

    • How can ad-hoc coordination/communication between algorithms and strangers work?

    • What lessons from zero-shot learning can be transferred to multi-agent systems?

    • Is natural language the best possible protocol for zero-shot cooperation?

    • How can theory of mind be used in zero-shot cooperation?

    • Which role can reasoning play in deep emergent communication?


We invite submissions from researchers both inside and outside the machine learning community in the following areas:

    • multi-agent communication

    • grounding emergent protocols to natural language

    • compositionality in emergent/natural languages

    • linguistic generalization

    • learning cognitive skills through language

    • language evolution

    • deep multi-agent learning


Submission Deadline: 16th October 2020 at 23:59 Anywhere on Earth

Submission Website: https://cmt3.research.microsoft.com/EmeCom2020

Submission Format: either extended abstracts (4 pages) or full papers (8 pages) anonymously using:

    • either the style guidelines for NeurIPS 2020 (found here), or

    • the style guidelines for ICLR 2021 (found here) or

    • the style guidelines for AAMAS 2021 (found here).

References and appendix should be appended into the same (single) PDF document, and do not count towards the page count. The camera-ready version of the papers should use the NeurIPS 2020 style guidelines.

All accepted papers will be part of a virtual poster session. There will be no formal publication of workshop proceedings. However, the accepted papers will be made available online on the workshop website as non-archival reports to allow submissions to future conferences/journals.


For any queries, reach out to us at emecomworkshop@gmail.com