Autoencoder-augmented Neuroevolution for Visual Doom Playing

July 12, 2017 Β· Declared Dead Β· πŸ› IEEE Conference on Computational Intelligence and Games

πŸ‘» CAUSE OF DEATH: Ghosted
No code link whatsoever

"No code URL or promise found in abstract"

Evidence collected by the PWNC Scanner

Authors Samuel Alvernaz, Julian Togelius arXiv ID 1707.03902 Category cs.AI: Artificial Intelligence Cross-listed cs.NE Citations 62 Venue IEEE Conference on Computational Intelligence and Games Last Checked 3 months ago
Abstract
Neuroevolution has proven effective at many reinforcement learning tasks, but does not seem to scale well to high-dimensional controller representations, which are needed for tasks where the input is raw pixel data. We propose a novel method where we train an autoencoder to create a comparatively low-dimensional representation of the environment observation, and then use CMA-ES to train neural network controllers acting on this input data. As the behavior of the agent changes the nature of the input data, the autoencoder training progresses throughout evolution. We test this method in the VizDoom environment built on the classic FPS Doom, where it performs well on a health-pack gathering task.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

πŸ“œ Similar Papers

In the same crypt β€” Artificial Intelligence

Died the same way β€” πŸ‘» Ghosted