Research map / Reinforcement learning / Visual and representation RL
Wasserstein Unsupervised Reinforcement Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use Wasserstein distance intrinsic rewards rather than mutual-information skill rewards | Unsupervised skill learning on HalfCheetah · HalfCheetah | WD: 13.01 $\pm$ 2.89 -> 39.21 $\pm$ 0.44 | Helped |
| Use Wasserstein distance intrinsic rewards rather than mutual-information skill rewards | Unsupervised skill learning on Ant · Ant | WD: 4.50 $\pm$ 0.27 -> 8.57 $\pm$ 0.39 | Helped |
| Use Wasserstein distance intrinsic rewards rather than mutual-information skill rewards | Unsupervised skill learning on Humanoid · Humanoid | WD: 279.94 $\pm$ 27.98 -> 556.96 $\pm$ 27.78 | Helped |
| Use amortized projected Wasserstein rewards instead of DIAYN-I pre-trained skills | Hierarchical policy · FreeRun navigation | reward: 15.06 $\pm$ 26.97 -> 125.56 $\pm$ 12.63 | Helped |
| Use amortized projected Wasserstein rewards instead of DADS-I pre-trained skills | Hierarchical policy · FreeRun navigation | reward: 109.78 $\pm$ 27.08 -> 125.56 $\pm$ 12.63 | Helped |
| Use amortized projected Wasserstein rewards instead of DIAYN-I pre-trained skills | Hierarchical policy · Ant navigation | reward: 35.00 $\pm$ 7.07 -> 100.00 $\pm$ 18.03 | Helped |
| Use amortized projected Wasserstein rewards instead of DADS-I pre-trained skills | Hierarchical policy · Ant navigation | reward: 46.00 $\pm$ 10.84 -> 100.00 $\pm$ 18.03 | Helped |
| Use minimum pairwise Wasserstein distance rather than mean pairwise distance as the reward (ablation) | WURL with multiple policies · FreeRun and TreeMaze | policy diversity | Helped |
| Add amortized stepwise credit assignment to projected Wasserstein rewards (ablation) | Projected Wasserstein distance policy learning · FreeRun | WD | Helped |
Closest papers by meaning
- Towards Principled Unsupervised Multi-Agent Reinforcement Learning (NeurIPS 2025 · similarity 0.90)
- Task Adaptation from Skills: Information Geometry, Disentanglement, and New Objectives for Unsupervised Reinforcement Learning (ICLR 2024 · similarity 0.90)
- Constrained Ensemble Exploration for Unsupervised Skill Discovery (ICML 2024 · similarity 0.89)
- Skill Disentanglement in Reproducing Kernel Hilbert Space (AAAI 2025 · similarity 0.89)
- Foundation Policies with Hilbert Representations (ICML 2024 · similarity 0.89)
- Disentangled Unsupervised Skill Discovery for Efficient Hierarchical Reinforcement Learning (NeurIPS 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.