View source

for Publications:Neuro-visual Control in the Quake II Game Engine

The action you have requested is limited to users in the group user.

You can view and copy the source of this page:

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.


Matt Parker and Bobby D. Bryant (2008).
Neuro-visual Control in the Quake II Game Engine.
''Proceedings of the 2008 International Joint Conference on Neural Networks (IJCNN’08)'', pp. 3827-3832. Piscataway, NJ: IEEE Press.

* Self-archived [http://nebl.cse.unr.edu/archive/papers/parker-2008-ijcnn.pdf PDF]

* Download the [[Software:Quake2AI|software]].

Return to Main Page.

Retrieved from "http://nebl.cse.unr.edu/wiki/Publications:Neuro-visual_Control_in_the_Quake_II_Game_Engine"