Trinkle23897 / VizDoom2018-Track1

Supplemental materials for "Playing FPS Games with Environment-Aware Hierarchical Reinforcement Learning", accepted by IJCAI'19

Home Page:https://www.ijcai.org/proceedings/2019/0482.pdf

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Supplemental Materials

Appendix

Including an experiment "Navigation Task".

Video Demos

Including 3 demos of ViZDoom single-player sceranios, performed by StarNet.

133.mp4

It first bypasses enemies and picks up some guns, after which it kills all of these enemies within 10 seconds, and finally successfully finishes the game.

56.mp4

It first picks resources, kills two monsters, picks guns, and kills all of enemies.

94.mp4

Almost the same as above.

About

Supplemental materials for "Playing FPS Games with Environment-Aware Hierarchical Reinforcement Learning", accepted by IJCAI'19

https://www.ijcai.org/proceedings/2019/0482.pdf