HalfCheetah-v-5
Emerging4papers using it
2026first seen
HalfCheetah-v5 is a benchmark environment used to evaluate the performance and stability of reinforcement learning algorithms, specifically in the context of training agents to navigate a simulated half-cheetah character.
Papers using HalfCheetah-v-5 (4)
- Not All Transitions Matter: Evidence from PPOVisualizing Latent Phase Structures in Locomotion Policies: A Multi-Environment Study with Temporal Feature ExtensionHindsight Preference Replay Improves Preference-Conditioned Multi-Objective Reinforcement LearningUncovering Latent Phase Structures and Branching Logic in Locomotion Policies: A Case Study on HalfCheetah