Evolving Communication in Embodied Agents: Theory, Methods, and Evaluation

  title={Evolving Communication in Embodied Agents: Theory, Methods, and Evaluation},
  author={Marco Mirolli and Stefano Nolfi},
  booktitle={Evolution of Communication and Language in Embodied Agents},
In this chapter we introduce the area of research that attempts to study the evolution of communication in embodied agents through adaptive techniques, such us artificial evolution. More specifically, we illustrate the theoretical assumptions behind this type of research, we present the methods that can be used to realize embodied and communicating artificial agents, and we discuss the main research challenges and the criteria for evaluating progresses in this field. 
Emergence of communication and language in evolving robots
This chapter illustrates how communication systems originate and evolve in a population of robots that adapt to a given task/environment and analysis of these synthetic experiments can help understand.
Evolution of Communication in Robots
This chapter contains sections titled: 8.2 Experimental Setup, 8.3 Results, and 8.4 Discussion, Supplemental Data, References.
On topic selection strategies in multi-agent naming game
This paper extends the language game model by proposing novel strategies for selecting topics, i.e. attracting agent's attention to different objects during the learning process, and demonstrates that the new strategies improve the overall convergence of the alignment process, limit agent's overall demand on memory, and scale with the increasing number of the interacting agents.
Cross-Population Semiosis in Multi-agent Systems
This research defines a settings for the cross-population semiosis model, which involves two or more mature populations set together in a common environment with a goal to align their predefined (or differently developed) lexicons, and introduces a new type of language game scenario and defines a set of specific measures that capture the dynamics of lexicon evolution in cross- Population semiosis.
Evolution of Communication-Based Collaborative Behavior in Homogeneous Robots
A method for role differentiation developed in previous work is applied to the evolution of teams of homogeneous robots in which role differentiation is based on a dedicated communication channel and it is shown that evolved solutions are considerably robust with respect to the number of robots composing the group.
Multi-participant Interaction in Multi-agent Naming Game
A meta-model of the Language Game is introduced that serves as a generalisation of the classical approach and facilitates better organisation and structuring of future research in the field.
ITALK - Integration and Transfer of Action and Language Knowledge in Robots
Whether a robot can be treated similar to an infant in an interaction is raised and the results confirm previous findings concerning the differences between ACI, AAI, and ARI and constitute an important empirical basis for making use of ostensive stimuli as social signals for tutoring behavior in social robotics.
Evolvable Mathematical Models: A New Artificial Intelligence Paradigm
NoiseWorld, an Artificial Life simulation in which interagent communication emerges and evolves from initially noncommunicating EMM-based agents, is developed, yielding significant insight into the emergence and subsequent evolution of cooperative communication.
Lights and Shadows in the Evolution of Language
James Hurford’s Language in the Light of Evolution two-volume project aims to bring together an up-to-date account of the synthesis of language in the light of evolution, going from meaningful mental


Emergence of communication in embodied agents evolved for the ability to solve a collective navigation problem
How evolving robots develop a non-trivial communication system and exploit different communication modalities and the possibility of co-adapting the robots’ individual and social/communicative behaviour plays a key role in the development of progressively more complex and effective individuals is indicated.
Emergence of communication in embodied agents: co-adapting communicative and non-communicative behaviours
In this paper, I discuss in which conditions a population of embodied and situated agents that have to solve problems that require co-operation might develop forms of ritualized interaction and
Social learning and cultural evolution in embodied and situated agents
Results demonstrate that social learning provides an adaptive advantage when individuals are allowed to learn socially from experienced individuals and individually and indicate how the iteration of a social and individual learning through out generations lead to a cumulative cultural evolutionary process.
Progress in the Simulation of Emergent Communication and Language
Although progress in this area has been substantial, it is able to identify some important areas for future research in the evolution of language, including the need for further computational investigation of key aspects of language such as open vocabulary and the more complex aspects of syntax.
Evolution of Communication
The two paradigms of communication are introduced, seen to be the extreme poles on a continuum, and it is concluded that traditional SP approaches are based on an ill-formed approach to communication.
Evolving Communication without Dedicated Communication Channels
  • M. Quinn
  • Computer Science, Biology
  • 2001
It is argued that functional, but non-communicative behaviours which evolve in the early stages of the simulation both make possible, and form the basis of, the communicative behaviour which subsequently evolves.
Social coordination and spatial organization: Steps towards the evolution of communication
Two sets of computational experiments are presented, one dealing with non-recursive coordination and the other with coordination of recursive actions, and one shows the inappropriateness of the metaphor of communication as an exchange of information.
Embedded neural networks: exploiting constraints
Understanding intelligence
Rolf Pfeifer and Christian Scheier provide a systematic introduction to this new way of thinking about intelligence and computers and derive a set of principles and a coherent framework for the study of naturally and artificially intelligent systems, or autonomous agents.
Power and the limits of reactive agents