Obstacle Tower

홈페이지 · 논문 20편

Obstacle Tower is a high fidelity, 3D, 3rd person, procedurally generated environment for reinforcement learning. An agent playing Obstacle Tower must learn to solve both low-level control and high-level planning problems in tandem while learning from pixels and a sparse reward signal. Unlike other benchmarks such as the Arcade Learning Environment, evaluation of agent performance in Obstacle Tower is based on an agent’s ability to perform well on unseen instances of the environment. Source: [Obstacle Tower: A Generalization Challenge in Vision, Control, and Planning](/paper/obstacle-tower-a-generalization-challenge-in)

3DEnvironment

벤치마크

General Reinforcement Learning on Obstacle Tower (No Gen) fixed 결과 2개

General Reinforcement Learning on Obstacle Tower (No Gen) varied 결과 2개

General Reinforcement Learning on Obstacle Tower (Strong Gen) fixed 결과 2개

General Reinforcement Learning on Obstacle Tower (Strong Gen) varied 결과 2개

General Reinforcement Learning on Obstacle Tower (Weak Gen) fixed 결과 2개

General Reinforcement Learning on Obstacle Tower (Weak Gen) varied 결과 2개