Research map / Reinforcement learning / Visual and representation RL
PlayVirtual: Augmenting Cycle-Consistent Virtual Trajectories for Reinforcement Learning
Tao Yu, Cuiling Lan, Wenjun Zeng, Mingxiao Feng, Zhizheng Zhang, Zhibo Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| augment SPR with cycle-consistent virtual trajectories | SPR on Atari-100k · Atari-100k | median HNS (%): 37.1 -> 47.2 | Improved |
| augment SPR† with cycle-consistent virtual trajectories | SPR† on DMControl-100k · DMControl-100k | median score: 728.0 -> 797.0 | Improved |
| augment SPR† with cycle-consistent virtual trajectories | SPR† on DMControl-500k · DMControl-500k | median score: 920.0 -> 935.0 | Improved |
| add future-state prediction (ablation) | SPR on Atari-100k · Atari-100k | median HNS (%): 33.4 -> 37.1 | Improved |
| add BDM without virtual-trajectory augmentation (ablation) | SPR on Atari-100k · Atari-100k | median HNS (%): 37.1 -> 38.4 | No clear effect |
| add cycle-consistent virtual trajectories (ablation) | Baseline+BDM on Atari-100k · Atari-100k | median HNS (%): 38.4 -> 47.2 | Improved |
| allow cycle loss to update the dynamics model as well as the encoder (ablation) | PlayVirtual-ND on Atari-100k · Atari-100k | median HNS (%): 44.0 -> 47.2 | Improved |
| set prediction steps to K=12 instead of K=9 (ablation) | PlayVirtual on Atari-100k · Atari-100k | median HNS (%): 47.2 -> 43.1 | Hurt |
| calculate cycle distance in projection space rather than latent space (ablation) | PlayVirtual on Atari-100k · Atari-100k | median HNS (%): 44.8 -> 47.2 | Improved |
| remove cycle-loss weight warmup (ablation) | PlayVirtual on DMControl-100k · DMControl-100k | median score: 749.5 -> 797.0 | Hurt |
| apply cycle constraint at every trajectory step instead of only the end step (ablation) | PlayVirtual on DMControl-100k · DMControl-100k | median score: 781.0 -> 797.0 | Hurt |
| increase virtual trajectories from M=20 to M=30 (ablation) | PlayVirtual on DMControl-100k · DMControl-100k | median score: 806.0 -> 792 | Hurt |
Later papers that compared with this method
4 later accepted papers report PlayVirtual as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Return-Critic: Bridging Goal Discrepancy for Efficient Visual Reinforcement Learning (ICML 2026 · better on 2 of 2 shared results)
- Value-Consistent Representation Learning for Data-Efficient Reinforcement Learning (AAAI 2023 · better on 0 of 1 shared result)
- PLASTIC: Improving Input and Label Plasticity for Sample Efficient Reinforcement Learning (NeurIPS 2023 · better on 1 of 1 shared result)
- Mask-based Latent Reconstruction for Reinforcement Learning (NeurIPS 2022 · better on 5 of 6 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Value-Consistent Representation Learning for Data-Efficient Reinforcement Learning (AAAI 2023 · similarity 0.88)
- For SALE: State-Action Representation Learning for Deep Reinforcement Learning (NeurIPS 2023 · similarity 0.87)
- Towards Principled Representation Learning from Videos for Reinforcement Learning (ICLR 2024 · similarity 0.86)
- Bridging State and History Representations: Understanding Self-Predictive RL (ICLR 2024 · similarity 0.85)
- The Surprising Ineffectiveness of Pre-Trained Visual Representations for Model-Based Reinforcement Learning (NeurIPS 2024 · similarity 0.85)
- Temporal Distance-aware Transition Augmentation for Offline Model-based Reinforcement Learning (ICML 2025 · similarity 0.85)
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.
