Obstacle Tower: A Generalization Challenge in Vision, Control, and Planning

The rapid pace of recent research in AI has been driven in part by the presence of fast and challenging simulation environments. These environments often take the form of games; with tasks ranging from simple board games, to competitive video games... (read more)

PDF Abstract

Datasets


Introduced in the Paper:

Obstacle Tower

Mentioned in the Paper:

OpenAI Gym Arcade Learning Environment VizDoom
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
General Reinforcement Learning Obstacle Tower (No Gen) fixed PPO Score 5 # 2
General Reinforcement Learning Obstacle Tower (No Gen) fixed RNB Score 7 # 1
General Reinforcement Learning Obstacle Tower (No Gen) varied RNB Score 4.8 # 1
General Reinforcement Learning Obstacle Tower (No Gen) varied PPO Score 1 # 2
General Reinforcement Learning Obstacle Tower (Strong Gen) fixed PPO Score 0.6 # 1
General Reinforcement Learning Obstacle Tower (Strong Gen) fixed RNB Score 0.6 # 1
General Reinforcement Learning Obstacle Tower (Strong Gen) varied PPO Score 0.6 # 2
General Reinforcement Learning Obstacle Tower (Strong Gen) varied RNB Score 0.8 # 1
General Reinforcement Learning Obstacle Tower (Weak Gen) fixed RNB Score 1 # 2
General Reinforcement Learning Obstacle Tower (Weak Gen) fixed PPO Score 1.2 # 1
General Reinforcement Learning Obstacle Tower (Weak Gen) varied RNB Score 3.4 # 1
General Reinforcement Learning Obstacle Tower (Weak Gen) varied PPO Score 0.8 # 2

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet