Reinforcement learning with replacing eligibility traces
Machine Learning - Special issue on reinforcement learning
Introduction to Reinforcement Learning
Introduction to Reinforcement Learning
Evolutionary Robotics: The Biology,Intelligence,and Technology
Evolutionary Robotics: The Biology,Intelligence,and Technology
Metalearning and neuromodulation
Neural Networks - Computational models of neuromodulation
Policy Invariance Under Reward Transformations: Theory and Application to Reward Shaping
ICML '99 Proceedings of the Sixteenth International Conference on Machine Learning
Adaptive Behavior - Animals, Animats, Software Agents, Robots, Adaptive Systems
Evolutionary Function Approximation for Reinforcement Learning
The Journal of Machine Learning Research
Evolution, learning, and instinct: 100 years of the baldwin effect
Evolutionary Computation
Co-evolution of Shaping Rewards and Meta-Parameters in Reinforcement Learning
Adaptive Behavior - Animals, Animats, Software Agents, Robots, Adaptive Systems
Potential-based shaping and Q-value initialization are equivalent
Journal of Artificial Intelligence Research
Evolutionary Development of Hierarchical Learning Structures
IEEE Transactions on Evolutionary Computation
Co-evolution of Rewards and Meta-parameters in Embodied Evolution
Creating Brain-Like Intelligence
Emergence of Different Mating Strategies in Artificial Embodied Evolution
ICONIP '09 Proceedings of the 16th International Conference on Neural Information Processing: Part II
Neuroevolution based on reusable and hierarchical modular representation
ICONIP'08 Proceedings of the 15th international conference on Advances in neuro-information processing - Volume Part I
Self-organization and specialization in multiagent systems through open-ended natural evolution
EvoApplications'12 Proceedings of the 2012t European conference on Applications of Evolutionary Computation
Learning potential functions and their representations for multi-task reinforcement learning
Autonomous Agents and Multi-Agent Systems
| Hi-index | 0.00 |
In this article we propose a framework for performing embodied evolution with a limited number of robots, by utilizing time-sharing in subpopulations of virtual agents hosted in each robot. Within this framework, we explore the combination of within-generation learning of basic survival behaviors by reinforcement learning, and evolutionary adaptations over the generations of the basic behavior selection policy, the reward functions, and metaparameters for reinforcement learning. We apply a biologically inspired selection scheme, in which there is no explicit communication of the individuals芒聙聶 fitness information. The individuals can only reproduce offspring by mating芒聙聰a pair-wise exchange of genotypes芒聙聰and the probability that an individual reproduces offspring in its own subpopulation is dependent on the individual芒聙聶s 芒聙聵芒聙聵health,芒聙聶芒聙聶 that is, energy level, at the mating occasion. We validate the proposed method by comparing it with evolution using standard centralized selection, in simulation, and by transferring the obtained solutions to hardware using two real robots.