Timezone: »

Emergent Communication Workshop
Jakob Foerster · Igor Mordatch · Angeliki Lazaridou · Kyunghyun Cho · Douwe Kiela · Pieter Abbeel

Sat Dec 09 08:00 AM -- 06:30 PM (PST) @ S4
Event URL: https://sites.google.com/site/emecom2017/ »

Communication is one of the most impressive human abilities. The question of how communication arises has been studied for many decades, if not centuries. However, due to the computational and representational limitations, in the past problem-settings had to be restricted to low dimensional, simple observation spaces. With the rise of deep reinforcement learning methods, this question can now be studied in complex multi-agent settings, which has lead to flourishing activity in the area over the last two years. In these settings agents can learn to communicate in grounded multi-modal environments and rich communication protocols emerge.

However, the recent research has been largely disconnected from the study of emergent communication in other fields and even from work done on this topic in previous decades. This workshop will provide a forum for a variety of researchers from different fields (machine learning, game-theory, linguistics, cognitive science, and programming languages) interested in the question of communication and emergent language to exchange ideas.


Sat 8:50 a.m. - 9:00 a.m.
Opening Remarks (talk through day ahead, schedule, etc)
Sat 9:00 a.m. - 9:30 a.m.
"Language in context" (Invited Talk)
Noah Goodman
Sat 9:30 a.m. - 10:00 a.m.
"Communication via Physical Action" (Invited Talk)
Anca Dragan
Sat 10:00 a.m. - 10:30 a.m.
Invited Talk 3 (Invited Talk)
Pushmeet Kohli
Sat 10:30 a.m. - 11:00 a.m.
Coffee Break (Break)
Sat 11:00 a.m. - 12:15 p.m.

All speakers from the morning session and further invited panelists.

Jürgen Schmidhuber, Noah Goodman, Anca Dragan, Pushmeet Kohli, Dhruv Batra
Sat 12:15 p.m. - 2:15 p.m.
Poster Session and Lunch (Poster session)
Sat 2:15 p.m. - 3:00 p.m.
  • Vehicle Communication Strategies for Simulated Highway Driving. Cinjon Resnick*, NYU; Ilia Kulikov, NYU; Kyunghyun Cho, New York University; Jason Weston, FAIR

  • Multiagent Bidirectionally-Coordinated Nets: Emergence of Human-level Coordination in Learning to Play StarCraft Combat Games. Ying Wen*, UCL; Jun Wang, UCL

  • A two dimensional decomposition approach for matrix completion through gossip. Mukul Bhutani*, Amazon; Bamdev Mishra, Amazon.com

  • Multi-Agent Compositional Communication Learning from Raw Visual Input. Edward Choi*, Georgia Institute of Technology; Angeliki Lazaridou, DeepMind; Nando de Freitas, DeepMind

  • Analyzing Language Learned by an Active Question Answering Agent. Jannis Bulian*, Google; Christian Buck, Google; Massimiliano Ciaramita, Google; Wojciech Gajewski, Google; Andrea Gesmundo, Google; Neil Houlsby, Google; Wei Wang, Google

  • MACE: Structured Exploration via Deep Hierarchical Coordination. Stephan Zheng*, Caltech; Yisong Yue, Caltech

  • Multi-agent Communication by Bi-directional Recurrent Neural Networks. Ying Wen*, UCL; Yaodong Yang, University College London; Jun Wang, UCL

Cinjon Resnick, Ying Wen, Stephan Zheng, Mukul Bhutani, Edward Choi
Sat 3:00 p.m. - 3:30 p.m.
Coffee Break + Poster Presentation (Break and Posters)
Sat 3:30 p.m. - 4:00 p.m.
  • Modeling Opponent Hidden States in Deep Reinforcement Learning. Roberta Raileanu*, NYU; Arthur Szlam, FAIR; Rob Fergus, FAIR

  • Emergent Communication through Negotiatio.n. Kris Cao*, University of Cambridge; Angeliki Lazaridou, DeepMind; Stephen Clark, University of Cambridge/Deepmind; Joel Z Leibo, DeepMind; Marc Lanctot, Deepmind; Karl Tuyls, Deepmind

-Natural Language Does Not Emerge 'Naturally' in Multi-Agent Dialog. Satwik Kottur*, Carnegie Mellon University; Satwik Kottur; José M.F. Moura; Stefan Lee; Dhruv Batra (EMNLP 2017)

  • Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols. Serhii Havrylov*, Institute for Language, Cognition and Computation, the University of Edinburgh; Ivan Titov, Institute for Language, Cognition and Computation, the University of Edinburgh (NIPS 2017)

  • An unbroken evolutionary pathway to symbolic communication. Paul Grouchy*, University of Toronto; Paul Grouchy, Gabriele M. T. D’Eleuterio; Morten H. Christiansen; Hod Lipson (Nature Scientific Reports)

Roberta Raileanu, Satwik Kottur, Paul Grouchy
Sat 4:00 p.m. - 4:30 p.m.
"From Democritus to Signaling Networks" (Invited Talk)
Brian Skyrms
Sat 4:30 p.m. - 5:00 p.m.
"Language Emergence as Boundedly Optimal Control" (Invited Talk)
Satinder Singh
Sat 5:00 p.m. - 6:15 p.m.

Brian Skyrms, Satinder Singh, Jacob Andreas

Brian Skyrms, Satinder Singh, Jacob Andreas
Sat 6:15 p.m. - 6:30 p.m.
Closing Remarks (closing remarks)

Author Information

Jakob Foerster (University of Oxford)

Jakob Foerster received a CIFAR AI chair in 2019 and is starting as an Assistant Professor at the University of Toronto and the Vector Institute in the academic year 20/21. During his PhD at the University of Oxford, he helped bring deep multi-agent reinforcement learning to the forefront of AI research and interned at Google Brain, OpenAI, and DeepMind. He has since been working as a research scientist at Facebook AI Research in California, where he will continue advancing the field up to his move to Toronto. He was the lead organizer of the first Emergent Communication (EmeCom) workshop at NeurIPS in 2017, which he has helped organize ever since.

Igor Mordatch (OpenAI)
Angeliki Lazaridou (DeepMind)
Kyunghyun Cho (NYU)

Kyunghyun Cho is an associate professor of computer science and data science at New York University and a research scientist at Facebook AI Research. He was a postdoctoral fellow at the Université de Montréal until summer 2015 under the supervision of Prof. Yoshua Bengio, and received PhD and MSc degrees from Aalto University early 2014 under the supervision of Prof. Juha Karhunen, Dr. Tapani Raiko and Dr. Alexander Ilin. He tries his best to find a balance among machine learning, natural language processing, and life, but almost always fails to do so.

Douwe Kiela (Facebook AI Research)
Pieter Abbeel (University of California, Berkley & OpenAI)

More from the Same Authors