Autoencoder-augmented Neuroevolution For Visual Doom Playing

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.

Stats

• citations33
• S2 citations—
• github stars0
• HF likes0
• heat score11.49
• arxiv keyalvernaz2017autoencoder

Related papers

• Deep Reinforcement Learning For Doom Using Unsupervised Auxiliary Tasks (2018)0.00
• Vizdoom Competitions: Playing Doom From Pixels (2018)13.79
• Agents That Listen: High-throughput Reinforcement Learning With Multiple Sensory Systems (2021)8.09
• Vizdoom: DRQN With Prioritized Experience Replay, Double-q Learning, & Snapshot Ensembling (2018)8.60
• Playing Atari With Six Neurons (2018)0.00
• VINE: An Open Source Interactive Data Visualization Tool For Neuroevolution (2018)0.00
• Do We Need Transformers To Play FPS Video Games? (2025)0.00
• Neuroevolution Of Recurrent Architectures On Control Tasks (2023)4.52