Les méthodes évolutionnistes peinent à mettre au point des artefacts complexes lorsque la fitness est insuffisante pour guider explicitement le processus. Supposant que la complexité des êtres vivants provient en partie de la multiplicité des pressions sélectives, nous proposons la création de telles pressions pour l'évolution de réseaux de neurones à l'aide d'algorithmes évolutionnistes multiobjectifs. Nous commençons par décrire comment des hypothèses sur des étapes intermédiaires peuvent être exploitées à l'aide d'une optimisation multiobjectif. Nous envisageons ensuite plusieurs méthodes multiobjectifs pour maintenir la diversité des comportements des solutions. Enfin, nous montrons que les exaptations peuvent être favorisées via des pr...
The application of multi-objective optimisation to evolutionary robotics is receiving increasing att...
Designing optimal soft modular robots is difficult, due to non-trivial interactions between morpholo...
In machine learning, the fitness function is an incredibly important part of defining and learning h...
The goal of this research is to develop a behavior oriented design methodology dedicated to mobile a...
This paper investigates the use of a multi-objective approach for evolving artificial neural network...
A modular approach to neural behavior control of autonomous robots is presented. It is based on the ...
<div><p>The application of multi-objective optimisation to evolutionary robotics is receiving increa...
The application of multi-objective optimisation to evolutionary robotics is receiving increas-ing at...
This paper is concerned with artificial evolution of neuro-controllers with adaptive synapses for au...
Neuromodulation is a biologically-inspired technique that can adapt the per-connection learning rate...
This research explores a new approach of using a multi-objective evolutionary algorithm (MOEA) to ev...
Evolutionary robotics is concerned with the design of intelligent systems with life-like properties ...
The utilization of a multi-objective approach for evolving artificial neural networks that act as th...
We propose a method for evolving neural network controllers robust with respect to variations of the...
The application of multi-objective optimisation to evolutionary robotics is receiving increasing att...
The application of multi-objective optimisation to evolutionary robotics is receiving increasing att...
Designing optimal soft modular robots is difficult, due to non-trivial interactions between morpholo...
In machine learning, the fitness function is an incredibly important part of defining and learning h...
The goal of this research is to develop a behavior oriented design methodology dedicated to mobile a...
This paper investigates the use of a multi-objective approach for evolving artificial neural network...
A modular approach to neural behavior control of autonomous robots is presented. It is based on the ...
<div><p>The application of multi-objective optimisation to evolutionary robotics is receiving increa...
The application of multi-objective optimisation to evolutionary robotics is receiving increas-ing at...
This paper is concerned with artificial evolution of neuro-controllers with adaptive synapses for au...
Neuromodulation is a biologically-inspired technique that can adapt the per-connection learning rate...
This research explores a new approach of using a multi-objective evolutionary algorithm (MOEA) to ev...
Evolutionary robotics is concerned with the design of intelligent systems with life-like properties ...
The utilization of a multi-objective approach for evolving artificial neural networks that act as th...
We propose a method for evolving neural network controllers robust with respect to variations of the...
The application of multi-objective optimisation to evolutionary robotics is receiving increasing att...
The application of multi-objective optimisation to evolutionary robotics is receiving increasing att...
Designing optimal soft modular robots is difficult, due to non-trivial interactions between morpholo...
In machine learning, the fitness function is an incredibly important part of defining and learning h...