|
Origins of Language Publications
The following publications have been made by the origins of
language group. Most of the publications can be downloaded in
(compressed) PostScript format.
- PhD Thesis: Autonomous Formation of Concepts and Communication.
Edwin D. de Jong, June 2000. Available
online.
-
De Jong,
E.D. (2000). Attractors in
the Development of Communication. [PDF]. J.-A. Meyer, A. Berthoz,
D. Floreano, H. Roitblat, and S. Wilson (Eds). SAB2000 Proceedings
Supplement Book (to appear). Honolulu, Hawaii: International Society
for Adaptive Behavior. This is a shorter version of AI-MEMO 00-02 [PDF].
Abstract: The development of communication in a population of
agents is viewed as the behavior of a dynamical system. A deterministic
communication system is shown experimentally to have point attractors that
correspond to perfect communication. However, the determinism required for
the presence of point attractors impedes the development of communication.
A corresponding attractor type for the stochastic system is defined, and
the existence of these attractors in a stochastic version of the system is
demonstrated.
-
De Jong,
E.D. (1999). Analyzing the
Evolution of Communication from a Dynamical Systems Perspective. In
Proceedings of the European Conference on Artificial Life ECAL'99, 689-693.
©Springer-Verlag
LNCS,
Berlin. Click here
for a Gzipped PostScript version. An extended version appeared as AI-MEMO (.ps.gz) of the VUB AI
Lab.
Abstract: We study the evolution of communication where
concepts are developed individually by agents and relations between
concepts and forms (words, signals) are learned through interaction with
other agents. By constructing concepts based on experience with the same
environment, agents develop similar conceptual systems. Concepts represent
situations in the environment. The system of associations between forms
and meanings is viewed as a dynamical system. This paper presents first
results with investigating the phase space of the system. The analysis
contributed to understanding the interaction between association strengths
of different agents and of different meanings.
- De Jong,
E.D. (1999). Autonomous
Concept Formation. In Proceedings of the International Joint
Conference on Artificial Intelligence IJCAI'99, pp. 344-349.
Abstract: A model for the formation of situation
concepts is described. A characteristic of this form of concept
formation is that it does not require instructive feedback. This renders
it suitable for concept formation by autonomous agents. It is
experimentally demonstrated that situation concepts constructed
independently by several agents can convey useful information between
agents through a learned system of communication. A relation was found
between the development of the learned system of communication and the
duration of the situations.
-
De Jong,
E.D. and L. Steels (1999). Generation and selection of
sensory channels. In Evolutionary Image Analysis, Signal Processing
and Telecommunications
First European Workshops, EvoIASP'99 and EuroEcTel'99 Joint Proceedings,
pp. 90-100. Göteborg, Sweden, May 1999. ©Springer-Verlag
LNCS,
Berlin. Click here for
a Gzipped PostScript version.
Abstract:
Sensory channels determine the way an agent views the world. We
investigate the question of how sensory channels may be autonomously
constructed using generation and selection. The context is the
discrimination of geometric shapes. In a first experiment, elements of a
solution were attributed fitness based on the part of the problem they
solved. In two subsequent experiments, cooperation between elements was
respectively required and encouraged by means of a fitness function which
only rewards complete solutions. Differences between the approaches are
discussed, and generation and selection is concluded to provide a
successful mechanism for the autonomous construction of sensory
channels.
-
De Jong,
E.D. (1999). Coordination
Developed by Learning from Evaluations. To appear in the collected
papers of the VIM workshops, ©Springer-Verlag
LNAI,
Berlin. Click here
for a Gzipped PostScript version. An earlier version appeared in the Notes
of the VIM'97 Workshop
on Collaboration between human and artificial societies. Universita' di
Salerno.
Abstract: This paper reports on research into the origins of
communication and coordination. Several problems with defining
communication and coordination are
noted. A research methodology is described that circumvents these problems.
The methodology is used in an experiment concerning the development of
coordination.
The aim of the experiment is to see whether a learning agent can use
coordination signals, which represent evaluations of its behavior,
to learn to coordinate its actions in an unknown environment. The task is
a pursuit problem where four agents are needed to capture a randomly
moving prey. One of these agents adapts its behavior based on the
coordination signals it receives from the three other agents. The
development of coordination increased the capture rate in this pursuit
problem from an initial 5% to 93%. Thus, in combination with a general
learning mechanism, coordination signals may be sufficient for the
development of coordination.
- De Jong,
E.D. (1998). The
Development of a Lexicon Based on Behavior. In: Proceedings of the
Tenth Dutch Conference on Artificial Intelligence NAIC'98 (in press). Click
here
for a Gzipped PostScript version.
Abstract: This paper investigates whether a group of agents
may develop a common lexicon relating words to situations by a process
of self-organization. Each agent independently decides which situations
are useful to distinguish, based on its experience with the environment.
It then starts to associate signals with each situation. The agents adapt
their own associations based on the signals they received from other
agents.
The system is monitored using measures which reflect the development of
the lexicons over time. The result of the distributed activities of the
agents is that a coherent shared lexicon emerges linking signals to
situations.
-
De Jong, E.D. and P.
Vogt (1998) How
Should a Robot Discriminate Between Objects? A comparison between two
methods. Proceedings of the Fifth International Conference of The Society
for Adaptive Behavior SAB'98
Abstract In order to discriminate among the different objects
in its environment, an agent may develop a primitive notion of concepts
based on the sensor data it receives. In this paper, this phenomenon is
investigated by having software agents play discrimination games with the
sensor data of autonomous robots. We have compared the Simple Prototype
method and the Adaptive Subspace method. Both methods achieve high
discrimination-success
rates. The Adaptive Subspace method accomplishes this with a converging
and relatively low number of categories. The purpose of these
discrimination
games is to serve as a basis for lexicon formation experiments. From the
experiments in this paper, we conclude that the Adaptive Subspace method
is more attractive for discrimination games.
-
Steels, L. and P.
Vogt (1997) Grounding
adaptive language games in robotic agents. In Harvey, P. and P.
Husbands
(eds.) (1997) Proceedings of the 4th European Conference on Artificial
Life 97.
Abstract The paper addresses the question how a group of
physically
embodied robotic agents may originate meaning and language through adaptive
language games. The main principles underlying the approach are sketched
as well as the steps needed to implement these principles on physical
agents.
Some experimental results based on this implementation are
presented.
-
Vogt, P. (1998) Perceptual
grounding in robots. In (Birk, A. and J. Demiris eds.) Proceedings
of the 6th European Workshop on Learning Robots 1997. Lecture Notes
on Artificial Intelligence. Springer-Verlag 1998.
Abstract This paper reports on an experiment in which robotic
agents are able to ground objects in their environment using low-level
sensors. The reported experiment is part of a larger experiment, in which
autonomous agents ground an adaptive language through self-organization.
Grounding is achieved by the implementation of the hypothesis that meaning
can be created using mechanisms like feature generation and
self-organization.
The experiments were carried out to investigate how agents may construct
features in order to learn to discriminate objects from each other. Meaning
is formed to give semantic value to the language, which is also created
by the agents in the same experiments. From the experimental results we
can conclude that the robots are able to ground meaning in this
self-organizing
manner. This paper focuses on the meaning creation and will only discuss
the language formation very briefly. The paper sketches the tested
hypothesis,
the experimental set-up and experimental results.
-
Vogt, P. (1998) The
evolution of a lexicon and meaning in robotic agents through
self-organization
AI-Memo 98-09. Presented at the 2nd International Conference on the
Evolution
of Language, London, April 6-9 1998
Abstract This paper discusses interdisciplinary experiments,
combining robotics and evolutionary computational linguistics. The goal
of the experiments is to investigate if robotic agents can originate a
language, in particular a lexicon. In the experiments two robots engage
in a series of so-called language games. Starting from the assumption that
the robots know how to communicate and are able to detect some sensory
information from the environment, the agents ground conceptual meaning
and develop a lexicon. The experiments show that the robots are able to
form a shared communication system. The paper investigates the influence
of using non-linguistic information in the formation of the lexicon, which
takes the form of pointing (1) to indicate the topic of the language game,
and (2) to give feedback on the outcome of the game.
Back to Top
|
