Abstract
Experiments to illustrate a novel methodology for
reinforcement learning in embodied physical agents are described.
A simulated legged robot is decomposed into structurebased
modules following the authors' EMBER principles of
local sensing, action and learning. The legs are individually
trained to 'walk' in isolation, and re-attached to the robot;
walking is then sufficiently stable that learning in situ can
continue. The experiments demonstrate the benefits of the
modular decomposition: state-space factorisation leads to faster
learning, in this case to the extent that an otherwise intractable
problem becomes learnable.
reinforcement learning in embodied physical agents are described.
A simulated legged robot is decomposed into structurebased
modules following the authors' EMBER principles of
local sensing, action and learning. The legs are individually
trained to 'walk' in isolation, and re-attached to the robot;
walking is then sufficiently stable that learning in situ can
continue. The experiments demonstrate the benefits of the
modular decomposition: state-space factorisation leads to faster
learning, in this case to the extent that an otherwise intractable
problem becomes learnable.
| Original language | English |
|---|---|
| Title of host publication | Procs 2005 IEEE Int Symposium on Computational Intelligence in Robotics and Automation |
| Subtitle of host publication | CIRA 2005 |
| Publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| Pages | 365-372 |
| ISBN (Print) | 0-7803-9355-4 |
| Publication status | Published - 2005 |
| Event | 2005 IEEE Int Symposium of Computational Intelligence in Robotics & Automation - Espoo, Finland Duration: 27 Jun 2005 → 30 Jun 2005 |
Conference
| Conference | 2005 IEEE Int Symposium of Computational Intelligence in Robotics & Automation |
|---|---|
| Country/Territory | Finland |
| City | Espoo |
| Period | 27/06/05 → 30/06/05 |