# Learning Cross-Domain Correspondence for Control with Dynamics Cycle-Consistency

arXiv: https://arxiv.org/abs/2012.09811 · HTML: https://mlsanity.com/paper/2012.09811/

## Key facts

- Cited by 43 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Dec 2020.

## Built on by (most cited first, 30 of 43)

- [Learning by Watching: Physical Imitation of Manipulation Skills from Human Videos](https://mlsanity.com/paper/2101.07241/) (2021)
- [Humanoid Policy ~ Human Policy](https://mlsanity.com/paper/2503.13441/) (2025)
- [PlayVirtual: Augmenting Cycle-Consistent Virtual Trajectories for Reinforcement Learning](https://mlsanity.com/paper/2106.04152/) (2021)
- [Cross-Domain Policy Adaptation via Value-Guided Data Filtering](https://mlsanity.com/paper/2305.17625/) (2023)
- [Polybot: Training One Policy Across Robots While Embracing Variability](https://mlsanity.com/paper/2307.03719/) (2023)
- [Cross-Embodiment Dexterous Grasping with Reinforcement Learning](https://mlsanity.com/paper/2410.02479/) (2024)
- [Cross-Domain Policy Adaptation by Capturing Representation Mismatch](https://mlsanity.com/paper/2405.15369/) (2024)
- [Cross-Embodiment Robot Manipulation Skill Transfer using Latent Space Alignment](https://mlsanity.com/paper/2406.01968/) (2024)
- [Neural Unbalanced Optimal Transport via Cycle-Consistent Semi-Couplings](https://mlsanity.com/paper/2209.15621/) (2022)
- [A Comprehensive Survey of Cross-Domain Policy Transfer for Embodied Agents](https://mlsanity.com/paper/2402.04580/) (2024)
- [Policy Transfer across Visual and Dynamics Domain Gaps via Iterative Grounding](https://mlsanity.com/paper/2107.00339/) (2021)
- [Learning Feasibility to Imitate Demonstrators with Different Dynamics](https://mlsanity.com/paper/2110.15142/) (2021)
- [Policy Adaptation from Foundation Model Feedback](https://mlsanity.com/paper/2212.07398/) (2022)
- [ILBiT: Imitation Learning for Robot Using Position and Torque Information based on Bilateral Control with Transformer](https://mlsanity.com/paper/2401.16653/) (2024)
- [Learning Shadow Correspondence for Video Shadow Detection](https://mlsanity.com/paper/2208.00150/) (2022)
- [xTED: Cross-Domain Adaptation via Diffusion-Based Trajectory Editing](https://mlsanity.com/paper/2409.08687/) (2024)
- [Cross Domain Robot Imitation with Invariant Representation](https://mlsanity.com/paper/2109.05940/) (2021)
- [Weakly Supervised Correspondence Learning](https://mlsanity.com/paper/2203.00904/) (2022)
- [CUP: Critic-Guided Policy Reuse](https://mlsanity.com/paper/2210.08153/) (2022)
- [AnyBody: A Benchmark Suite for Cross-Embodiment Manipulation](https://mlsanity.com/paper/2505.14986/) (2025)
- [Dual-Robust Cross-Domain Offline Reinforcement Learning Against Dynamics Shifts](https://mlsanity.com/paper/2512.02486/) (2025)
- [Learning from Imperfect Demonstrations via Adversarial Confidence Transfer](https://mlsanity.com/paper/2202.02967/) (2022)
- [Learning Action Translator for Meta Reinforcement Learning on Sparse-Reward Tasks](https://mlsanity.com/paper/2207.09071/) (2022)
- [Cross Domain Policy Transfer with Effect Cycle-Consistency](https://mlsanity.com/paper/2403.02018/) (2024)
- [Similarity-based Knowledge Transfer for Cross-Domain Reinforcement Learning](https://mlsanity.com/paper/2312.03764/) (2023)
- [Invariance Co-training for Robot Visual Generalization](https://mlsanity.com/paper/2512.05230/) (2025)
- [Learning ABCs: Approximate Bijective Correspondence for isolating factors of variation with weak supervision](https://mlsanity.com/paper/2103.03240/) (2021)
- [Extraneousness-Aware Imitation Learning](https://mlsanity.com/paper/2210.01379/) (2022)
- [Out-of-Dynamics Imitation Learning from Multimodal Demonstrations](https://mlsanity.com/paper/2211.06839/) (2022)
- [Action-Conditioned World Model for Goal Plane Probe Guidance in Robotic Ultrasound](https://mlsanity.com/paper/2607.21918/) (2026)

## Abstract (arXiv)

At the heart of many robotics problems is the challenge of learning correspondences across domains. For instance, imitation learning requires obtaining correspondence between humans and robots; sim-to-real requires correspondence between physics simulators and the real world; transfer learning requires correspondences between different robotics environments. This paper aims to learn correspondence across domains differing in representation (vision vs. internal state), physics parameters (mass and friction), and morphology (number of limbs). Importantly, correspondences are learned using unpaired and randomly collected data from the two domains. We propose textitdynamics cycles that align dynamic robot behavior across two domains using a cycle-consistency constraint. Once this correspondence is found, we can directly transfer the policy trained on one domain to the other, without needing any additional fine-tuning on the second domain. We perform experiments across a variety of problem domains, both in simulation and on real robot. Our framework is able to align uncalibrated monocular video of a real robot arm to dynamic state-action trajectories of a simulated arm without paired data. Video demonstrations of our results are available at: https://sjtuzq.github.io/cycledynamics.html .