Combining Attention and Value Maps
We present an approach where we combine attention with value maps for the purpose of acquiring a decision-making policy for multiple concurrent goals. The former component is essential for dealing with an uncertain and open environment while the latter offers a general model for building decision-making systems based on reward information. We discuss the multiple goals policy acquisition problem and justify our approach. We provide simulation results that support our solution.
Unable to display preview. Download preview PDF.
- 1.Sutton, R., Barto, A.: Reinforcement Learning: An Introduction, 4th edn. MIT Press, Cambridge (2002)Google Scholar
- 3.Taylor, J.G.: Attentional movement: the control basis for Consciousness. Soc. Neuroscience Abstracts 26, 2231, #893.3 (2000)Google Scholar
- 4.Kasderidis, S., Taylor, J.G.: Attentional Agents and Robot Control. International Journal of Knowledge-based and Intelligent Systems 8, 69–89 (2004)Google Scholar
- 5.Kasderidis, S., Taylor, J.G.: Attention-based Learning. In: International Joint Conference on Neural Networks (IJCNN 2004), Budapest, 25-29 July, pp. 525–531 (2004)Google Scholar