Pith. sign in

REVIEW 4 major objections 4 minor 32 references

World Translation claims that the hidden factors that make simulation diverge from reality can be read backward from each observed transition, and that translating these features across domains yields more accurate real-world dynamics predi

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-01 15:49 UTC pith:PJEZI2Y5

load-bearing objection Genuinely new idea for bypassing history-based hidden-state inference in sim-to-real dynamics, but the empirical claims need more transparency and the identifiability claim is narrower than advertised. the 4 major comments →

arxiv 2607.18154 v1 pith:PJEZI2Y5 submitted 2026-07-20 cs.RO

World Translation: Minimizing Sim-to-Real Gap with Backward Dynamics Extraction and Unpaired Domain Translation

Xinchen Yao , Leixin Chang , Hua Chen This is my paper
classification cs.RO
keywords sim-to-real transferlearned dynamics modelspartial observabilitybackward dynamics extractionunpaired domain translationcycle consistencylatent dynamics featureslegged locomotion
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper is trying to establish that the unobservable physical factors behind sim-to-real divergence—sudden contacts, payload motion, external forces—do not need to be recovered from observation history, as existing learned-dynamics models assume. They can instead be read backward from the outcome of each individual step, because the effect of a hidden cause is written in the transition itself. The extracted feature is then treated as dynamics content that can be translated from simulation to reality with cycle-consistent unpaired translation, preserving what the hidden cause is while changing how the domain renders it. If this holds, policies can be trained in a simulator that has been corrected to follow real-world dynamics, with the largest gains in exactly the situations where history is least informative, such as a sudden collision.

Core claim

The central claim is that hidden variable h_t and domain characteristic c can be encoded into a latent dynamics feature z_t = ω(h_t, c) extracted backward from the observed transition (o_t,a_t,o_{t+1}), sufficient to reconstruct the next observation. Extraction runs from outcome to cause, so no informative history is required. Because the same hidden cause yields different transitions in each domain, a cycle-consistent map G_{S→R} translates z_t while preserving hidden-variable content. Trained on unpaired sim and real transitions and used to overwrite simulator states, the method beats history-based state-space models in high-uncertainty tasks (Go2 payload MSE 0.148 vs 0.298) and real-robot

What carries the argument

The load-bearing object is the backward-extracted dynamics feature z_t = ω(h_t, c), produced by a variational encoder that maps (o_t, a_t, o_{t+1}) to a latent distribution. The decoder reconstructs o_{t+1} from (o_t, a_t, z_t); an adversarial blind decoder keeps z_t from becoming a copy of o_{t+1}, a domain classifier forces z_t to encode domain identity, and feature-wise modulation biases the decoder to rely on z_t rather than domain cues in (o_t,a_t). The second half is a cycle-consistent translator G_{S→R} with reverse G_{R→S} acting on z_t in latent space, keeping the discriminator from separating domains using the observation context. In deployment, a simulator transition is encoded, t

Load-bearing premise

The whole method rests on the assumption that the hidden variable's effect is identifiable from a single observed transition (o_t,a_t,o_{t+1}); when two different hidden states produce the same transition, the extracted feature is ambiguous and the translation cannot transfer the true cause.

What would settle it

On a system where two distinct hidden configurations (e.g., a payload ball at rest in two different positions inside a box) produce the same observed transition, measure World Translation's single-step prediction error and a probe's R² for recovering the configuration from the transition. If error is no better than Direct Prediction's, or R² is near zero, the backward-extraction premise fails in a regime the method is designed to improve.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Learned dynamics models can drop the history-recoverability assumption; sudden unannounced events become learnable from their own outcome.
  • Unpaired simulated and real transition data become usable for dynamics transfer, removing the requirement of matched pairs with identical hidden states.
  • Policies can be trained directly on corrected simulator rollouts where every next state follows translated real dynamics, with state overwriting keeping trajectories physical.
  • The advantage over history-based models should widen as tasks involve more abrupt hidden-variable events, and shrink when hidden causes leave no signature in the outcome.
  • Real-robot fine-tuning on the Go2 quadruped with payload lowers angular-velocity tracking error compared with no adaptation and with domain randomization.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The identifiability premise is approximate even on the paper's own evidence: Table I recovers only R²=0.73–0.78 of the true force from a transition, and §VI.A concedes continuous forces leave weaker signatures; in settings with even lower R², backward extraction will approach the behavior of history-free prediction.
  • Pairing the translated state target with an action-correction scheme, an extension the authors name in §VI.C, could keep the simulator's internal buffers consistent and likely extend the stable rollout horizon beyond the step-30 point where state overwriting accumulates error.
  • Because the translation is unpaired, the map can be updated incrementally as new real transitions arrive, suggesting an online adaptation loop that does not require collecting matched simulation data.
  • If the latent factorization into h_t and c is as clean as the classifier probes suggest, the same representation could double as a system-identification readout, estimating real friction or motor gains from the translated feature.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper proposes World Translation, a sim-to-real dynamics-alignment method that combines backward dynamics extraction with unpaired domain translation. The central idea is to encode unobservable dynamics information (hidden variables and domain characteristics) from an already-observed transition (o_t, a_t, o_{t+1}) into a latent dynamics feature z_t, rather than inferring it from observation history as in conventional world models. A CycleGAN-style translator then maps these features between simulation and reality, and the translated feature is decoded into a real-domain next observation. The authors instantiate the framework with a VAE, a blind-decoder adversarial regularizer, a domain classifier, FiLM conditioning, and cycle-consistent translation. Experiments on G1 humanoid, Go2 quadruped, and R5 manipulator tasks compare single-step prediction error against DirectPred, RawSim, ResDyn, and RSSM, and a real-robot Go2 experiment with a payload is reported. The paper claims consistent gains, largest when unobservable factors are not recoverable from history.

Significance. The core idea is genuinely novel and potentially valuable: extracting hidden-variable information from the transition outcome instead of from history, and translating that latent representation across domains, addresses a real limitation of history-based learned dynamics models under sudden contact or payload disturbances. The paper also includes useful diagnostics (representation probes and latent-vs-observation translation ablation) and a real-robot pilot. If the empirical validation is made rigorous, the approach could be a meaningful step for learned dynamics in sim-to-real transfer. However, the current evidence is not sufficient to establish the breadth of the central claim, primarily because the identifiability assumption is only partially supported and the evaluation protocol lacks held-out data, pairing details, and statistical variance.

major comments (4)
  1. [§III.A, Eq. (1); Table I; §VI.A] The central claim rests on the identifiability assumption in §III.A: 'requiring only that the effect of h_t be identifiable from the observed transition.' The paper's own probe results in Table I show that the full transition (o_t, a_t, o_{t+1}) predicts the normalized external force with R²=0.73–0.78, so roughly a quarter of the hidden-variable variance is not recoverable. §VI.A concedes that continuous forces leave weaker signatures. This is not a mere caveat: if two distinct h_t produce the same observed transition, the encoder's z_t is an aliased estimate of h_t, and cycle-consistent translation will preserve that aliasing. The §I claim that backward extraction is 'more broadly reliable than history-based methods' is therefore stronger than the evidence. A direct identifiability test is needed—e.g., construct hidden-state pairs that yield identical transitions and measure whether the
  2. [§V-A.3, Fig. 4 (teacher-forcing protocol)] The paper does not specify how source and target transitions are paired when computing single-step error. The protocol says only that the model receives simulator transitions and error is measured against 'ground-truth' target transitions. Under the paper's unpaired-data premise (§III.C), a randomly chosen target transition does not share h_t with the simulator transition; in that case the reported MSE conflates prediction error with hidden-variable mismatch. If, instead, target transitions are selected so that h_t is shared, that pairing must be described and justified. This is essential for interpreting all quantitative comparisons (e.g., Go2 Payload 0.148 vs. RSSM 0.298).
  3. [§V-A.3, §V-B, §V-D; Fig. 4; Tables I–IV] Neither the simulation experiments nor the real-robot experiment report error bars or variance across seeds/trials. The real-robot result appears to be a single deployment. Without repeated trials, it is impossible to assess whether the claimed gains (e.g., R5 Force 0.401 vs. RSSM 0.515; Go2 tracking 0.1786 vs. 0.2475) are significant. The paper should provide at least 3–5 seeds for each simulation task and multiple independent trials for the robot experiment.
  4. [§V-A.3 and §V-B (train/eval split)] The evaluation may be circular for the target domain. The encoder, decoder, and translation models are trained on target-domain transitions sampled from D_R, and the same type of target data is used to compute the reported prediction errors. The text does not state that the evaluation transitions are held out from training, so the 'prediction error' may be reconstruction error rather than generalization error. A held-out split and a precise description of how target evaluation transitions are selected are required.
minor comments (4)
  1. [Fig. 2] The taxonomy box 'Absorbed inf' appears truncated; likely should read 'Absorbed in f'. Please correct.
  2. [§VI.C] The text refers to 'Appendix Table II' for hyperparameter transfer, but no appendix is included in the manuscript. Add the table or fix the reference.
  3. [Fig. 4 (left)] The diagram has duplicated 'e' labels near the target trajectory; clarify which quantity is being measured and against which prediction.
  4. [Eq. (8), Algorithm 1] The blind-decoder loss is described as adversarial, but the text does not fully specify gradient flow: the algorithm detaches z_t when updating ψ, yet the encoder's maximization of L_blind is not explicitly shown in Algorithm 1's update of φ, θ, ξ. Please clarify the exact stop-gradient/update recipe.

Circularity Check

0 steps flagged

No significant circularity: the derivation chain is self-contained and no prediction reduces by construction to a fitted input.

full rationale

World Translation's derivation chain is self-contained. The backward-extraction encoder is trained with a reconstruction objective on transitions (Eq. 11), and at inference it encodes simulator outcomes and translates the resulting features to the real domain; the target real transition is not fed into the encoder at prediction time. The VAE/CycleGAN training objectives are standard learning losses rather than fitted constants renamed as predictions. The paper's core assumption—'requiring only that the effect of h_t be identifiable from the observed transition' (§III.A)—is a stated condition, not a conclusion derived from itself; Table I's R²=0.73–0.78 and §VI.A's admission that continuous forces leave weaker signatures constrain the scope of the claim without making it circular. The representation probes (Tables I–II) are trained on the paper's own data and lack explicit train/test splits, which is a methodological self-referentiality concern, but it is not definitional circularity: the reported accuracies are not 1 by construction. There are no load-bearing self-citations (the reference list contains no works by the present authors), no imported uniqueness theorem, and no ansatz smuggled in via citation. The real-robot Go2 deployment provides an external benchmark. Hence the derivation does not reduce to its own inputs.

Axiom & Free-Parameter Ledger

6 free parameters · 5 axioms · 1 invented entities

The framework rests on identifiability of hidden variables from single transitions, approximate physical coverage by the simulator, and an unproven assumption that cycle consistency preserves hidden-variable semantics. These are stated explicitly in §III.A and §VI.A.

free parameters (6)
  • β (KL regularization weight) = not reported
    Hand-chosen loss weight in Eq. (11) balancing reconstruction against KL regularization.
  • λ_b (blind penalty weight) = not reported
    Hand-chosen weight for adversarial blind-decoder penalty in Eq. (8)/(11).
  • σ_b (blind temperature) = not reported
    Temperature controlling saturation scale of blind penalty in Eq. (8).
  • λ_c (domain classifier weight) = not reported
    Hand-chosen weight for domain classification loss in Eq. (9)/(11).
  • λ_cyc (cycle consistency weight) = not reported
    Hand-chosen cycle-consistency weight in §IV-B.
  • λ_id (identity loss weight) = not reported
    Hand-chosen identity loss weight in §IV-B.
axioms (5)
  • domain assumption The effect of hidden variable h_t is identifiable from a single transition (o_t, a_t, o_{t+1}).
    §III.A states identifiability is required; the method fails when two hidden states produce the same transition.
  • domain assumption The simulator captures the relevant physics at least approximately; phenomena entirely absent from the simulator cannot be translated.
    §VI.A explicitly acknowledges unmodeled phenomena are unrecoverable.
  • domain assumption Cycle consistency over unpaired dynamics features preserves hidden-variable semantics while transferring domain characteristics.
    §III.C/IV.B propose this; no proof is provided, only empirical probe results (Tables I-II).
  • domain assumption The decoder uses domain information from z_t rather than from (o_t, a_t) despite leakage.
    §IV.A Challenge 2; FiLM and classifier reduce but do not guarantee this; the paper says 'no theoretical guarantee'.
  • domain assumption Standard VAE and CycleGAN training assumptions hold, including balanced discriminators and stable hyperparameters across domains.
    The method relies on adversarial training stability; the paper acknowledges instability as a practical challenge in §VI.C.
invented entities (1)
  • dynamics feature z_t no independent evidence
    purpose: Latent code that is supposed to carry hidden-variable and domain-characteristic information; central to translation.
    Model-internal construct; probe experiments (force R², domain classification) are validation on the paper's own data, not an external falsifiable handle.

pith-pipeline@v1.3.0-alltime-deepseek · 11735 in / 12751 out tokens · 165172 ms · 2026-08-01T15:49:18.155690+00:00 · methodology

0 comments
read the original abstract

The gap between simulation and reality remains a fundamental challenge in deploying simulation-trained robotic policies in the real world. Real-to-sim methods narrow this gap from the real side, learning transition dynamics from real data to build a more realistic digital world. Learned dynamics models are their dominant instance. Such methods, however, face a partial observability problem: the same observation may branch to different transitions due to unobservable factors. Existing methods assume these factors can be recovered from observation history. However, this may fail whenever observation history is uninformative, such as a sudden contact event with no prior warning. To address this limitation, we propose \textit{World Translation}, which exploits a complementary strength of simulators and learned dynamics. Simulators are deterministic but physically imperfect, while learned models are accurate but underdetermined under partial observability. Rather than predicting transitions forward from history, we extract the unobservable dynamics information backward from an observed transition, then translate this feature across simulation and reality as an unpaired domain-translation problem that preserves dynamics content while transferring domain style. Experiments across humanoid, quadruped, and manipulator platforms show that our method achieves more accurate dynamics modeling than baselines, with the largest gains when unobservable factors cannot be recovered from observation history. Real-robot deployment on Go2 quadruped confirms improved policy transfer.

Figures

Figures reproduced from arXiv: 2607.18154 by Hua Chen, Leixin Chang, Xinchen Yao.

Figure 1
Figure 1. Figure 1: World Translation bridges the sim-to-real gap by extracting [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Taxonomy of unobservable factors. Conventional dynamics models [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Simulation environments. Left: G1 humanoid on irregular terrain. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Dynamics modeling evaluation. Left: teacher-forcing protocol where the model predicts translated observations [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Multi-step rollout on Go2 high-ht task. (a) Divergence rate (fraction of rollouts where base height or velocity exceeds physical bounds). (b) Prediction MSE. RSSM and RawSim show zero divergence but RSSM MSE remains above RawSim; World Translation remains stable through horizon 30 with only ∼2% divergence at horizon 50. prediction quality [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: Training collapse in observation-level translation. Discriminator [PITH_FULL_IMAGE:figures/full_fig_p006_7.png] view at source ↗
Figure 6
Figure 6. Figure 6: UMAP visualization of dynamics features on the R5 manipulator [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗
Figure 8
Figure 8. Figure 8: Real-world deployment of the Go2 quadruped with the payload [PITH_FULL_IMAGE:figures/full_fig_p007_8.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

32 extracted references · 5 linked inside Pith

  1. [1]

    Learning agile and dynamic motor skills for legged robots,

    J. Hwangbo, J. Lee, A. Dosovitskiy, D. Bellicoso, V . Tsounis, V . Koltun, and M. Hutter, “Learning agile and dynamic motor skills for legged robots,”Science Robotics, vol. 4, no. 26, p. eaau5872, 2019

    2019

  2. [2]

    Learning quadrupedal locomotion over challenging terrain,

    J. Lee, J. Hwangbo, L. Wellhausen, V . Koltun, and M. Hutter, “Learning quadrupedal locomotion over challenging terrain,”Science robotics, vol. 5, no. 47, p. eabc5986, 2020

    2020

  3. [3]

    Dextrah- rgb: Visuomotor policies to grasp anything with dexterous hands,

    R. Singh, A. Allshire, A. Handa, N. Ratliff, and K. Van Wyk, “Dextrah- rgb: Visuomotor policies to grasp anything with dexterous hands,” arXiv preprint arXiv:2412.01791, 2024

    Pith/arXiv arXiv 2024

  4. [4]

    Domain randomization for transferring deep neural networks from simulation to the real world,

    J. Tobin, R. Fong, A. Ray, J. Schneider, W. Zaremba, and P. Abbeel, “Domain randomization for transferring deep neural networks from simulation to the real world,” in2017 IEEE/RSJ international con- ference on intelligent robots and systems (IROS). IEEE, 2017, pp. 23–30

    2017

  5. [5]

    Sim-to- real transfer of robotic control with dynamics randomization,

    X. B. Peng, M. Andrychowicz, W. Zaremba, and P. Abbeel, “Sim-to- real transfer of robotic control with dynamics randomization,” in2018 IEEE international conference on robotics and automation (ICRA). IEEE, 2018, pp. 3803–3810

    2018

  6. [6]

    Solving rubik’s cube with a robot hand,

    I. Akkaya, M. Andrychowicz, M. Chociej, M. Litwin, B. McGrew, A. Petron, A. Paino, M. Plappert, G. Powell, R. Ribas,et al., “Solving rubik’s cube with a robot hand,”arXiv preprint arXiv:1910.07113, 2019

    Pith/arXiv arXiv 1910

  7. [7]

    System identification,

    L. Ljung, “System identification,” inSignal analysis and prediction. Springer, 1998, pp. 163–173

    1998

  8. [8]

    Closing the sim-to-real loop: Adapting simula- tion randomization with real world experience,

    Y . Chebotar, A. Handa, V . Makoviychuk, M. Macklin, J. Issac, N. Ratliff, and D. Fox, “Closing the sim-to-real loop: Adapting simula- tion randomization with real world experience,” in2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019, pp. 8973–8979

    2019

  9. [9]

    Analysis of randomization effects on sim2real transfer in reinforcement learning for robotic manipulation tasks,

    J. Josifovski, M. Malmir, N. Klarmann, B. L. ˇZagar, N. Navarro- Guerrero, and A. Knoll, “Analysis of randomization effects on sim2real transfer in reinforcement learning for robotic manipulation tasks,” in2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2022, pp. 10 193–10 200

    2022

  10. [10]

    Learning robotic policy with imagined transition: Mitigating the trade-off between robustness and optimality,

    W. Xiao, S. Lyu, Z. Gong, R. Wang, and D. Wang, “Learning robotic policy with imagined transition: Mitigating the trade-off between robustness and optimality,”arXiv preprint arXiv:2503.10484, 2025

    arXiv 2025

  11. [11]

    Learning latent dynamics for planning from pixels,

    D. Hafner, T. Lillicrap, I. Fischer, R. Villegas, D. Ha, H. Lee, and J. Davidson, “Learning latent dynamics for planning from pixels,” in International conference on machine learning. PMLR, 2019, pp. 2555–2565

    2019

  12. [12]

    Dream to control: Learning behaviors by latent imagination,

    D. Hafner, T. Lillicrap, J. Ba, and M. Norouzi, “Dream to control: Learning behaviors by latent imagination,” inInternational Conference on Learning Representations (ICLR), 2020

    2020

  13. [13]

    Mastering diverse domains through world models,

    D. Hafner, J. Pasukonis, J. Ba, and T. Lillicrap, “Mastering diverse domains through world models,”arXiv preprint arXiv:2301.04104, 2023

    Pith/arXiv arXiv 2023

  14. [14]

    Td-mpc2: Scalable, robust world models for continuous control,

    N. Hansen, H. Su, and X. Wang, “Td-mpc2: Scalable, robust world models for continuous control,”arXiv preprint arXiv:2310.16828, 2023

    Pith/arXiv arXiv 2023

  15. [15]

    Sim-to- real transfer with neural-augmented robot simulation,

    F. Golemo, A. A. Taiga, A. Courville, and P.-Y . Oudeyer, “Sim-to- real transfer with neural-augmented robot simulation,” inConference on Robot Learning. PMLR, 2018, pp. 817–828

    2018

  16. [16]

    Neuralsim: Augmenting differentiable simulators with neural net- works,

    E. Heiden, D. Millard, E. Coumans, Y . Sheng, and G. S. Sukhatme, “Neuralsim: Augmenting differentiable simulators with neural net- works,” in2021 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2021, pp. 9474–9481

    2021

  17. [17]

    Residual physics learning and system identification for sim-to-real transfer of policies on buoyancy assisted legged robots,

    N. Sontakke, H. Chae, S. Lee, T. Huang, D. W. Hong, and S. Hal, “Residual physics learning and system identification for sim-to-real transfer of policies on buoyancy assisted legged robots,” in2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2023, pp. 392–399

    2023

  18. [18]

    Sim-to-real: Learning agile locomotion for quadruped robots,

    J. Tan, T. Zhang, E. Coumans, A. Iscen, Y . Bai, D. Hafner, S. Bo- hez, and V . Vanhoucke, “Sim-to-real: Learning agile locomotion for quadruped robots,” inRobotics: Science and Systems (RSS), 2018

    2018

  19. [19]

    Dropo: Sim-to-real transfer with offline domain randomization,

    G. Tiboni, K. Arndt, and V . Kyrki, “Dropo: Sim-to-real transfer with offline domain randomization,”Robotics and Autonomous Systems, vol. 166, p. 104432, 2023

    2023

  20. [20]

    Active domain randomization,

    B. Mehta, M. Diaz, F. Golemo, C. J. Pal, and L. Paull, “Active domain randomization,” inConference on Robot Learning. PMLR, 2020, pp. 1162–1176

    2020

  21. [21]

    Pilco: A model-based and data-efficient approach to policy search,

    M. Deisenroth and C. E. Rasmussen, “Pilco: A model-based and data-efficient approach to policy search,” inProceedings of the 28th International Conference on machine learning (ICML-11), 2011, pp. 465–472

    2011

  22. [22]

    Deep rein- forcement learning in a handful of trials using probabilistic dynamics models,

    K. Chua, R. Calandra, R. McAllister, and S. Levine, “Deep rein- forcement learning in a handful of trials using probabilistic dynamics models,”Advances in neural information processing systems, vol. 31, 2018

    2018

  23. [23]

    Day- dreamer: World models for physical robot learning,

    P. Wu, A. Escontrela, D. Hafner, P. Abbeel, and K. Goldberg, “Day- dreamer: World models for physical robot learning,” inConference on robot learning. PMLR, 2023, pp. 2226–2240

    2023

  24. [24]

    Sim- to-real of soft robots with learned residual physics,

    J. Gao, M. Y . Michelis, A. Spielberg, and R. K. Katzschmann, “Sim- to-real of soft robots with learned residual physics,”IEEE Robotics and Automation Letters, 2024

    2024

  25. [25]

    Grounded action transformation for robot learning in simulation,

    J. Hanna and P. Stone, “Grounded action transformation for robot learning in simulation,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 31, no. 1, 2017

    2017

  26. [26]

    Reinforced grounded action transformation for sim-to-real transfer,

    H. Karnan, S. Desai, J. P. Hanna, G. Warnell, and P. Stone, “Reinforced grounded action transformation for sim-to-real transfer,” inIEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020

    2020

  27. [27]

    Asap: Aligning simulation and real-world physics for learning agile humanoid whole-body skills,

    T. He, J. Luo, W. He, T. Xiao, Y . Zhang, Z. Shi, S. Fidler, Y . Zhu, and L. Liu, “Asap: Aligning simulation and real-world physics for learning agile humanoid whole-body skills,” inRobotics: Science and Systems (RSS), 2025

    2025

  28. [28]

    Unpaired image- to-image translation using cycle-consistent adversarial networks,

    J.-Y . Zhu, T. Park, P. Isola, and A. A. Efros, “Unpaired image- to-image translation using cycle-consistent adversarial networks,” in Proceedings of the IEEE international conference on computer vision, 2017, pp. 2223–2232

    2017

  29. [29]

    Rl-cyclegan: Reinforcement learning aware simulation-to-real,

    K. Rao, C. Harris, A. Irpan, S. Levine, J. Ibarz, and M. Khansari, “Rl-cyclegan: Reinforcement learning aware simulation-to-real,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 11 157–11 166

    2020

  30. [30]

    Retinagan: An object-aware approach to sim-to-real transfer,

    D. Ho, K. Rao, Z. Xu, E. Jang, M. Khansari, and Y . Bai, “Retinagan: An object-aware approach to sim-to-real transfer,” in2021 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2021, pp. 10 920–10 926

    2021

  31. [31]

    Film: Visual reasoning with a general conditioning layer,

    E. Perez, F. Strub, H. De Vries, V . Dumoulin, and A. Courville, “Film: Visual reasoning with a general conditioning layer,” inProceedings of the AAAI conference on artificial intelligence, vol. 32, no. 1, 2018

    2018

  32. [32]

    Umap: Uniform manifold approximation and projection for dimension reduction,

    L. McInnes, J. Healy, and J. Melville, “Umap: Uniform manifold approximation and projection for dimension reduction,”arXiv preprint arXiv:1802.03426, 2018

    Pith/arXiv arXiv 2018