This paper explores how communication can be understood as an adaptation by agents to their environment. We model agents as recurrent neural networks. After arguing against systems which use discrete symbols to evolve communication, we supply our agents with a number of continuous communications channels. The agents use these channels to initiate real-valued signals which propagate through the environment, decaying over distance, perhaps being perturbed by environmental noise. Initially, the agents' signals appear random; over time, a structure emerges as the agents learn to communicate task-specific information about their environment. We demonstrate how different communication schemes can evolve for a task, and then discover a commonality between the schemes in terms of information passed between agents. From this we discuss what it means to communicate, and describe how a semantics emerges in the agents' signals relative to their task domain. Key Words: communication; evolutionary a...
|
3170
|
The mathematical theory of communication
– Shannon
- 1962
|
|
2331
|
Optimization by Simulated Annealing
– Kirkpatrick, Gelatt, et al.
- 1983
|
|
2229
|
A robust layered control system for a mobile robot
– Brooks
- 1986
|
|
1037
|
Intelligence without representation
– Brooks
- 1991
|
|
429
|
Connectionism and cognitive architecture: A critical analysis
– Fodor
- 1988
|
|
296
|
Recursive distributed representations
– Pollack
- 1990
|
|
255
|
Understanding Computers and Cognition
– Winograd, Flores
- 1987
|
|
181
|
The induction of dynamical recognizers
– Pollack
- 1991
|
|
177
|
Evolving dynamical neural networks for adaptive behavior. Adaptive Behavior 1(1):91-122
– Beer, Gallagher
- 1992
|
|
160
|
An Evolutionary Algorithm that Constructs Recurrent Neural Networks
– Angeline, Saunders, et al.
- 1993
|
|
140
|
The dynamics of collective sorting: Robot-like ants and ant-like robots
– Deneubourg, Goss, et al.
- 1991
|
|
118
|
D.: Seeing the Light: Artificial Evolution, Real Vision. From Animals to Animats 3
– Harvey, Husbands, et al.
- 1994
|
|
113
|
Cooperation without communication: Multiagent schema-based robot navigation
– Arkin
- 1992
|
|
113
|
Adaptive action selection
– Maes
- 1991
|
|
112
|
Designing Emergent Behaviors: From Local Interactions to Collective Intelligence
– Matari'c
- 1992
|
|
109
|
Evolution of communication in artificial organisms
– Werner, Dyer
- 1991
|
|
58
|
Synthetic ethology and the evolution of cooperative communication
– MacLennan, Burghardt
- 1993
|
|
58
|
Learning to behave socially
– Mataric
- 1994
|
|
57
|
Artificial Intelligence Meets Natural Stupidity
– McDermott
- 1981
|
|
54
|
A Case Study in the Behavior-oriented Design of Autonomous Agents,’ From Animals to Animats
– Steels
- 1994
|
|
52
|
Evolution as a theme in artificial life: The Genesys/Tracker system
– Jefferson, Collins, et al.
- 1991
|
|
50
|
Challenges for Complete Creature Architectures
– Brooks
- 1990
|
|
48
|
Antfarm: Towards simulated evolution
– Collins, Jefferson
- 1991
|
|
48
|
Evolving Artificial Intelligence
– Fogel
- 1992
|
|
43
|
Simulation of adaptive behavior in animats: Review and prospect
– Meyer, Guillot
- 1991
|
|
41
|
Dimensions of Communication and Social Organization in Multi-Agent Robotic Systems", From animals to animats 2
– Arkin, Hobbs
- 1992
|
|
37
|
Learning Behavior Networks from Experience
– Maes
- 1990
|
|
34
|
Towards robot cooperation
– McFarland
- 1994
|
|
30
|
A hierarchical classifier system implementing a motivationally autonomous animat
– Donnart
- 1994
|
|
28
|
From SAB90 to SAB94 : Four Years of Animat Research
– Meyer, Guillot
|
|
28
|
Evolution of herding behavior in artificial animals
– Werner, Dyer
|
|
25
|
Connectionist learning for control
– Barto
- 1990
|
|
24
|
Phase transitions in artificial intelligence systems
– Hogg, Huberman
- 1987
|
|
22
|
A dynamical systems perspective on autonomous agents
– Beer
- 1992
|
|
22
|
Fool’s gold: Extracting finite state machines from recurrent network dynamics
– Kolen
- 1994
|
|
18
|
D.: Representation for Artificial Organisms
– Collins, Jefferson
- 1991
|
|
12
|
Intelligence as Adaptive Behavior: An Experiment
– Beer
- 1990
|
|
12
|
Intentional Systems in Cognitive Ethology: the 'Panglossian paradigm' defended
– Dennett
- 1983
|
|
9
|
Signalling and territorial aggression: An investigation by means of synthetic behavioural ecology
– Bourcier, Wheeler
- 1994
|
|
7
|
Structural and behavioral evolution of recurrent networks
– Saunders, Angeline, et al.
- 1994
|
|
6
|
Behavioral dynamics of escape and avoidance: A neural network approach
– Schmajuk
- 1994
|
|
5
|
The Evolution of Strategies for Multi-agent Environments, Adaptive Behavior
– Grefenstette
- 1992
|
|
4
|
Modeling the role of cerebellum in prism adaptation
– Arbib, Schweighofer, et al.
- 1994
|
|
4
|
What kind of information processing is intelligence: A perspective on ai paradigms and a proposal
– Chandrasekaran
- 1990
|
|
4
|
The importance of leaky levels for behavior-based AI
– Saunders, Kolen, et al.
- 1994
|
|
1
|
Explorations of Back-Propagation and the Computational Capabilities of Recurrent Neural Networks
– Kolen
- 1994
|
|
1
|
Pollack The Evolution of Communication
– Saunders
- 1992
|
|
