The Community for Technology Leaders
Green Image
The ability to transfer knowledge learned in one environment in order to improve performance in a different environment is one of the hallmarks of human intelligence. Insights into human transfer learning help us to design computer-based agents that can better adapt to new environments without the need for substantial reprogramming. In this paper we study the transfer of knowledge by humans playing various scenarios in a graphically realistic urban setting which are specifically designed to test various levels of transfer. We determine the amount and type of transfer that is being performed based on the performance of human trained and untrained players. In addition, we use a graph-based relational learning algorithm to extract patterns from player graphs. These analyses reveal that indeed humans are transferring knowledge from one set of games to another and the amount and type of transfer varies according to player experience and scenario complexity. The results of this analysis help us understand the nature of human transfer in such environments and shed light on how we might endow computer-based agents with similar capabilities. The game simulator and human data collection also represent a significant testbed in which other AI capabilities can be tested and compared to human performance.
data mining, graph algorithms, transfer learning, games

L. B. Holder, G. M. Youngblood and D. J. Cook, "Graph-Based Analysis of Human Transfer Learning Using a Game Testbed," in IEEE Transactions on Knowledge & Data Engineering, vol. 19, no. , pp. 1465-1478, 2007.
80 ms
(Ver 3.3 (11022016))