PointMaze

PointMaze is a benchmark used to evaluate reinforcement learning agents in sparse-reward environments, focusing on their exploration efficiency and learning performance.

Papers using PointMaze (3)

Match or Replay: Self Imitating Proximal Policy Optimization2026PolicyFlow: Policy Optimization with Continuous Normalizing Flow in Reinforcement Learning2026Taming OOD Actions for Offline Reinforcement Learning: An Advantage-Based Approach2025