Mahaut, MatéoFranzon, FrancescaDessì, RobertoBaroni, Marco2023-07-312023-07-312023-07-31Mahaut M, Franzon F, Dessi R, Baroni M. Referential communication in heterogeneous communities of pre-trained visual deep networks. In: AAMAS 2023. 22nd International Conference on Autonomous Agents and Multiagent Systems; 2023 May 29-Jun 2; London, United Kingdom. [Richland]: IFAAMAS; 2023. p. 2619-21.9781450394321http://hdl.handle.net/10230/57743Comunicació presentada a la 22nd International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2023), celebrada a Londres del 29 de maig al 2 de juny de 2023.As large pre-trained image-processing neural networks are being embedded in autonomous agents such as self-driving cars or robots, the question arises of how such systems can communicate with each other about the surrounding world, despite their different architectures and training regimes. As a first step in this direction, we explore the task of referential communication in a community of state-of-the-art pre-trained visual networks, showing that they can develop a shared protocol to refer to a target image among a set of candidates. Such shared protocol, induced in a self-supervised way, can to some extent be used to communicate about previously unseen object categories. Finally, we show that a new neural network can learn the shared protocol developed in a community with remarkable ease, and the process of integrating a new agent into a community more stably succeeds when the original community includes a larger set of heterogeneous networks.application/pdfeng© 2023 International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved. https://www.ifaamas.org/Proceedings/aamas2023/info:eu-repo/semantics/conferenceObjectEmergent deep net communicationDeep visual netsMulti-agent communicationinfo:eu-repo/semantics/openAccess