Abstract | ||
---|---|---|
Backpropagation and neuroevolution are used in a Lamarckian evolution process to train a neural network visual controller for agents in the Quake II environment. In previous work, we hand-coded a non-visual controller for supervising in backpropagation, but hand-coding can only be done for problems with known solutions. In this research the problem for the agent is to attack a moving enemy in a visually complex room with a large central pillar. Because we did not know a solution to the problem, we could not hand-code a supervising controller; instead, we evolve a non-visual neural network as supervisor to the visual controller. This setup creates controllers that learn much faster and have a greater fitness than those learning by neuroevolution-only on the same problem in the same amount of time. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/CIG.2009.5286462 | Milano |
Keywords | Field | DocType |
complex room,non-visual neural network,non-visual controller,quake ii environment,neural network,lamarckian evolution process,quake ii,large central pillar,visual controller,greater fitness,known solution,human supervision,learning artificial intelligence,backpropagation,artificial neural networks,games,neuroevolution,visualization | Supervisor,Control theory,Visualization,Computer science,Simulation,Quake (series),Artificial intelligence,Backpropagation,Artificial neural network,Neuroevolution,Visual control,Machine learning | Conference |
ISBN | Citations | PageRank |
978-1-4244-4815-9 | 4 | 0.58 |
References | Authors | |
17 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Matt Parker | 1 | 56 | 7.28 |
Bobby D. Bryant | 2 | 58 | 6.70 |