Name
Abstract | ||
|---|---|---|
We explore the use of learning schemes in training and adapting performance on simple coordination tasks. The tasks are 1-D pole bal ancing. Several programs incorporating learning have already achieved this (1, S, 8): the problem is to move a cart along a short piece of track to at to keep a pole balanced on its end; the pole is hinged to the cart at its bottom, and the cart is moved either to the left or to the right by a force of constant magnitude. The form of the task considered here, after (3), involves a genuinely difficult credit-assignment prob lem. We use a learning scheme previously developed and analysed (1, 7) to achieve performance through reinforcement, and extend it to include changing and new requirements. For example, the length or mast of the pole can change, the bias of the force, its strength, and so on; and the system can be tasked to avoid certain regions altogether. In this way we explore the learning system's ability to adapt to changes and to profit from a selected training sequence, both of which are of obvious utility in practical robotics applications. The results described here were obtained using a computer sim ulation of the pole-balancing problem. A movie will be shown of the performance of the system under the various requirements and tasks. |
Year | Venue | Keywords |
|---|---|---|
1985 | IJCAI | genuinely difficult credit-assignment problem,constant magnitude,obvious utility,1-d pole balancing,new requirement,certain region,pole-balancing problem,selected training sequence,practical robotics application,computer simulation,profitability |
Field | DocType | ISBN |
Cart,Computer science,Pole balancing,Artificial intelligence,Reinforcement,Machine learning,Robotics | Conference | 0-934613-02-8 |
Citations | PageRank | References |
38 | 9.76 | 3 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Oliver G. Selfridge | 1 | 44 | 11.40 |
| Richard S. Sutton | 2 | 6100 | 1436.83 |
| Andrew G. Barto | 3 | 3937 | 829.22 |