Title :
Neuro-visual control in the Quake II game engine
Author :
Parker, Matt ; Bryant, Bobby D.
Author_Institution :
Dept. of Comput. Sci. & Eng., Univ. of Nevada, Reno, NV
Abstract :
The first-person-shooter Quake II is used as a platform to test neuro-visual control and retina input layouts. Agents are trained to shoot a moving enemy as quickly as possible in a visually simple environment, using a neural network controller with evolved weights. Two retina layouts are tested, each with the same number of inputs: first, a graduated density retina which focuses near the center of the screen and blurs outward; second, a uniform retina which focuses evenly across the screen. Results show that the graduated density retina learns more successfully than the uniform retina.
Keywords :
computer games; neurocontrollers; Quake II game engine; neural network controller; neuro-visual control; retina input layouts; Artificial intelligence; Cameras; Computational modeling; Engines; Focusing; Humans; Neural networks; Retina; Sensor arrays; Testing;
Conference_Titel :
Neural Networks, 2008. IJCNN 2008. (IEEE World Congress on Computational Intelligence). IEEE International Joint Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-1820-6
Electronic_ISBN :
1098-7576
DOI :
10.1109/IJCNN.2008.4634348