Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Video2Reward: Generating Reward Function from Videos for Legged Robot Behavior Learning

T0 review · 4 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper proposes Video2Reward, which turns videos of target behaviors into keypoint-coordinate texts that an LLM converts into reward functions, and shows this beats text-only reward generation by over 37.6% in human normalized score on…

desk verdict A promising video-to-reward idea whose headline results are undercut by metric reuse and missing variance. read the letter →

arxiv 2412.05515 v1 pith:PLSCFWQ7 submitted 2024-12-07 cs.RO cs.CVcs.LG

Runhao Zeng , Dingjie Zhou , Qiwei Liang , Junlin Liu , Hui Li , Changxin Huang , Jianqiang Li , Xiping Hu
show 1 more author
Fuchun Sun
This is my paper · ORCID
classification cs.ROcs.CVcs.LG
keywords video2rewardrewardfunctiongenerationlargelanguagemodelsleggedrobotskeypointtrajectoriesFastDTWreinforcementlearningbehaviorimitation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper proposes replacing textual task descriptions in LLM-based reward generation with video demonstrations: a pose estimator extracts joint keypoint trajectories from a video of the desired behavior, and these coordinate sequences are formatted as text along with the environment code and task description for an LLM to write an executable reward function. The paper's core claim is that this video-grounded reward generation produces legged-robot behaviors that are both more successful on the task and more visually similar to the target behavior than rewards generated from text alone, with human normalized scores over 37.6% higher than the leading text-based method on quadruped and over 11% higher on humanoid tasks. A second claim is that using the video again as a feedback signal — comparing simulated keypoint trajectories to the video trajectories with FastDTW and feeding the similarity scores back to the LLM — materially improves the reward over using video only as input. A sympathetic reader would care because reward shaping is currently the main bottleneck in legged-robot reinforcement learning, and this offers a path from raw internet video to trainable reward without expert manual tuning.

What carries the argument

The load-bearing mechanism is a video-to-text transforming module combined with a video-assisted feedback loop. The module uniformly samples frames, runs pose estimation (RTMpose for humans, HRNet for animals) to obtain J keypoint coordinates per frame, and formats the resulting trajectories as coordinate text; this text is concatenated with environment code, task description, and reward-design rules into the LLM prompt. For feedback, the trained policy is rolled out in simulation, its 3D keypoint trajectories are projected to 2D along the direction of motion, segmented into two-period chunks using autocorrelation, and each segment is compared with the video trajectory by FastDTW; the per-joint similarity scores become textual feedback appended to the prompt for the next round. The reward generation loop samples K candidate reward functions per round, trains each with PPO, selects the best by the maximum training success metric, and repeats for N rounds, so the video acts both as the initial specification and as the persistent performance signal.

What would settle it

Run Video2Reward on a reference video that depicts a locomotion pattern the LLM has not seen in its text training, e.g., a dog performing a pace gait, and compare the learned robot behavior against the same video using an independent motion-capture similarity metric that does not rely on the projection used in the paper; if the video-conditioned reward does not yield lower distance to the reference than a text-only baseline across multiple seeds, the central claim fails.

Watch

Extended reading notes

Core claim

The central discovery is that the motion information in a video, once reduced to keypoint coordinate trajectories, is a sufficiently precise learning objective for an LLM to produce reward functions that drive reinforcement-learned policies to reproduce the demonstrated behavior. The paper shows this in two ways: quantitatively, the learned policies achieve higher maximum training success and lower DTW trajectory distance to the reference video than policies trained with Eureka, Eureka with a text description of the gait, or the expert-designed IsaacGym rewards; qualitatively, the robots exhibit the alternating footfalls of an amble or the extended hind-leg propulsion of a run as shown in the input videos. The authors also demonstrate through ablation that the video-assisted feedback loop — in which the robot's projected 3D keypoint trajectories are compared with the video's 2D keypoint trajectories using FastDTW, and per-joint similarity scores are returned as text to the LLM — is responsible for a substantial share of the gain. The claim is conditional on the simulator and the chosen LLM, but the mechanism is presented as domain-general for legged behavior learning.

Load-bearing premise

The load-bearing premise is that 2D keypoint trajectories extracted from a video and the 2D projections of a simulated robot's 3D keypoints measure the same thing about a behavior, so that FastDTW scores and the resulting text feedback are meaningful.

Editorial extensions

If this is right

  • If the claim holds, switching the input video should change the learned gait without retraining the reward-design pipeline, enabling rapid re-purposing of one RL training setup across behaviors such as amble and run.
  • Reward functions produced from video should reduce the workload of manually designing dense rewards for new legged platforms, since the same video-to-text conversion and feedback loop can be applied whenever a reference movement video exists.
  • The video-assisted feedback is a stronger signal than text-only or reward-based feedback: even without feedback, video input beats text-only, and with feedback it surpasses expert-designed rewards on the tested tasks.
  • The DTW-score improvements imply that trajectories of individual joints, not just task-level metrics, align more closely to the reference, which is what makes the learned behavior look natural.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • One extension the authors do not test is sensitivity to camera viewpoint: projecting the robot's 3D trajectories to 2D along its direction of motion is a critical freedom, and a substantial change in the reference video's camera angle could alter similarity scores for the same behavior, suggesting a need for camera-aware or multi-view projection.
  • The same video-to-text and feedback loop could in principle apply to manipulation or whole-body tasks beyond legged locomotion, with the caveat that keypoint semantics must match between the video subject and the robot.
  • A testable refinement would be to normalize temporal dynamics before FastDTW comparison, since raw coordinate trajectories may conflate speed variations with postural differences; isolating phase-normalized pose could give a cleaner behavioral similarity signal.
  • As LLMs become increasingly multimodal, the explicit pose-extraction step might be replaced by direct video-conditioned reward generation, but the paper's evidence suggests that explicit keypoint text remains an effective and interpretable intermediate representation.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes Video2Reward, a method that uses videos depicting target behaviors to automatically generate reward functions for legged robot reinforcement learning. A video-to-text module extracts 2D keypoint trajectories from the video and feeds them, together with auxiliary textual context, to an LLM, which produces executable reward code. A video-assisted iterative refinement loop then compares the simulated robot's projected 3D keypoint trajectories with the video trajectories using FastDTW, converts the similarity scores into textual feedback, and prompts the LLM to refine the rewards. Experiments in IsaacGym on Anymal (amble and run) and Humanoid (run) tasks report that the method outperforms Eureka by 37.6% and 11.1% in human-normalized score, and also achieves lower DTW scores, indicating closer behavioral mimicry. The paper further presents ablations showing the contribution of video-assisted feedback and qualitative visualizations of learned gaits.

Significance. If the results are reliable, this work would be a meaningful step toward reducing manual reward engineering and enabling more controllable, video-specified behavior learning for legged robots. The proposed pipeline is straightforward and reproducible in principle, and the ablation study is a positive feature. However, the paper's central claims are weakened by two interconnected issues: the metric used to evaluate behavioral mimicry (DTW) is also the feedback signal used to optimize the reward via the LLM, and the trajectory comparison itself relies on an ill-defined 2D projection and an unspecified keypoint correspondence. The lack of multiple seeds and error bars further limits the strength of the quantitative conclusions. These concerns are specific and load-bearing; they affect both the headline performance numbers and the behavioral-mimicry claim, and they need to be addressed before the contribution can be considered established.

major comments (4)
  1. [Section 4.3, Algorithm 1, and Table 1] The DTW-based trajectory similarity is used both as the feedback signal d in Algorithm 1 (line 12) and as the evaluation metric for behavioral mimicry in Table 1. Since the LLM is explicitly prompted to reduce d during the refinement iterations, the lower DTW scores of the final policies are not independent evidence of closer behavioral mimicry; they may simply reflect optimization of the same objective. The authors should evaluate the learned policies with a held-out, independently defined similarity metric (for example, a learned pose-distance, a different projection, or a manual labeling of behavioral similarity) to support the claim that the method produces behaviors closer to the target videos.
  2. [Section 4.3] The projection of the simulated robot's 3D keypoint trajectories 'along the direction of motion' is not well-defined, and it is not matched to the camera geometry of the input video. The video keypoints are 2D projections under an unknown perspective camera, while the simulated keypoints are in 3D world coordinates; projecting them along an arbitrary axis can make the FastDTW similarity arbitrarily large or small. The paper also does not specify how the pixel-scale video coordinates are reconciled with the metric-scale simulation coordinates. The authors should specify the exact projection axis and normalization procedure, and show that the reported DTW scores and the feedback are robust to reasonable choices of projection direction.
  3. [Section 5.3 and 5.5] All experimental results are reported as single numbers without error bars, multiple random seeds, or statistical significance tests. Reinforcement learning in IsaacGym is stochastic with respect to initializations and sampling, so the headline claim of a 37.6% improvement over Eureka on the Anymal task is not supported without variance estimates. Please repeat the experiments with at least 3-5 seeds and report mean and standard deviation, and perform a significance test where appropriate.
  4. [Section 4.2 and Section 5.5 (Q1)] The paper compares keypoint trajectories extracted from the video with keypoints of the simulated robot, but it does not specify how the correspondence between, for example, a dog's nose, knees, and ankles in the video and the corresponding body parts of the Anymal robot is established. The HRNet and RTMpose keypoints follow a specific skeleton, while the simulated robot has its own kinematic structure. Without an explicit mapping, the per-keypoint DTW scores computed in Table 1 and the feedback used in Algorithm 1 are not well defined. The authors should provide the exact keypoint correspondence or use a common skeleton representation for both the video and the simulation.
minor comments (5)
  1. [Section 4.3 and Algorithm 1] The notation d is introduced as a scalar Fsim(T_hat, T), but earlier the paper computes per-joint similarity scores S_j; please clarify whether d is a scalar average or a vector/list and how it is formatted in the textual feedback.
  2. [Section 5.2] The 'Human' baseline is described only as a reward function designed by experts in IsaacGym; please specify which exact reward function is used and whether it is identical to the one used in the Eureka baseline for fair comparison.
  3. [Table 3 and Section 5.3] Human-normalized scores above 1.0 (e.g., 1.003 on Anymal and 1.180 on Humanoid) imply that the method surpasses the expert-designed reward; please comment on whether this is expected and whether the sparse-reward-based Hmts metric can be gamed by the policies.
  4. [Section 5.4] The video processing details, such as the frame-sampling interval and the normalization for moving cameras, are described only briefly and refer to supplementary material; please provide the essential details in the main text or ensure the supplementary material is included and accessible.
  5. [Section 5.5 and Conclusions] The paper claims to surpass 'state-of-the-art LLM-based reward generation methods' but only compares against Eureka; consider adding other baselines such as L2R or Text2Reward, or soften the claim to say that the method outperforms the evaluated baseline.

Circularity Check

1 steps flagged · score 6.0 of 10

The DTW-based behavioral-mimicry claim is partially circular: the same FastDTW score is fed back to the LLM as an optimization target and then reported as independent evidence of mimicry.

  1. fitted input called prediction [Algorithm 1 lines 11-14; Section 4.3 'Video-assisted reward function evaluation'; Section 5.3 'DTW score'; Table 1]
    "Calculate the similarity between T and ˆT: d = Fsim( ˆT , T ) ... Taux := Taux + Rn_best + sn_best + d ... The Dynamic Time Warping (DTW) score is employed as a metric to measure the similarity between behavioral trajectories."

    The same FastDTW function Fsim computes the d that is fed back to the LLM in line 14 and also produces the DTW scores reported in Table 1 as evidence that the learned behavior is closer to the target video. Because the LLM is explicitly prompted to reduce d and the best reward/policy is selected and refined through this feedback, Table 1 measures success at optimizing the method's own feedback signal rather than providing an independent test of behavioral mimicry. Without a held-out similarity metric, the conclusion that the learned behavior is 'closer to target behavior' is partly forced by construction.

full rationale

The core video-to-reward pipeline is not circular: the LLM could generate poor reward functions, and PPO training could fail, so the pipeline itself is not a tautology. However, the paper's headline behavioral-mimicry evidence is Table 1, and the DTW score used there is identical to the Fsim feedback signal injected into the LLM in Algorithm 1. This is a metric-reuse circularity: optimizing d and then reporting d as proof of success does not independently establish behavioral fidelity. The Hmts-based human normalized score is also used for selection in line 8, but since Eureka uses the same selection criterion, the +37.6% comparison is not invalidated by this reuse alone. The 2D projection 'along the direction of motion' raises a validity concern but is not a circularity per se. There is no load-bearing self-citation chain: the only same-author citation ([11]) is not invoked to justify the method. Overall, the partial circularity is concentrated in the DTW-based mimicry claim, warranting a score of 6 rather than higher; the central reward-generation result retains independent content.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The method does not introduce new physical entities or fitted physical constants. The free parameters are experimental hyperparameters and sampling choices. The key assumptions are domain assumptions about trajectory similarity, keypoint correspondence, and LLM capability. These are load-bearing but not backed by independent evidence in the paper.

free parameters (3)
  • Number of reward samples K and refinement iterations N = K=16, N=5
    Chosen to match Eureka's settings; no sensitivity analysis is reported, so performance may depend on these values.
  • Frame sampling interval for keypoint extraction = every 3 frames for humanoid, every 9 frames for Anymal
    Hand-chosen in Section 5.4 to capture 2-3 motion cycles; no ablation is provided for this choice.
  • Trajectory segmentation period = two periods per segment, detected by autocorrelation
    The period detection and segment length affect FastDTW similarity scores and thus the feedback; no robustness analysis is given.
assumptions (4)
  • domain assumption FastDTW similarity between projected 3D keypoint trajectories and 2D video keypoint trajectories is a valid measure of behavioral similarity.
    This is the core of the video feedback function in Section 4.3; if keypoint correspondences or the projection are not semantically meaningful, the feedback and the DTW evaluation lose validity.
  • domain assumption The LLM can generate executable and effective reward functions from coordinate trajectory text plus environment code.
    The whole pipeline in Equation 3 assumes the LLM can translate raw numeric trajectories into useful reward code; this is not proven independently of the experiments.
  • domain assumption Pose estimation keypoints from human and animal videos correspond to the robot's relevant joints.
    Section 4.2 uses RTMPose and HRNet keypoints as motion descriptors; if the correspondence to the simulated robot's body is loose, the learning target is ill-posed.
  • standard math Autocorrelation period detection and FastDTW are mathematically well-defined algorithms.
    The method relies on these standard tools for trajectory segmentation and similarity computation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Video2Reward: Generating Reward Function from Videos for Legged Robot Behavior Learning." pith.science (2026). https://pith.science/paper/PLSCFWQ7

@misc{pith2026241205515,
  author       = {Pith},
  title        = {Pith review of: Video2Reward: Generating Reward Function from Videos for Legged Robot Behavior Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PLSCFWQ7}},
  note         = {Machine review of arXiv:2412.05515}
}
read the original abstract

Learning behavior in legged robots presents a significant challenge due to its inherent instability and complex constraints. Recent research has proposed the use of a large language model (LLM) to generate reward functions in reinforcement learning, thereby replacing the need for manually designed rewards by experts. However, this approach, which relies on textual descriptions to define learning objectives, fails to achieve controllable and precise behavior learning with clear directionality. In this paper, we introduce a new video2reward method, which directly generates reward functions from videos depicting the behaviors to be mimicked and learned. Specifically, we first process videos containing the target behaviors, converting the motion information of individuals in the videos into keypoint trajectories represented as coordinates through a video2text transforming module. These trajectories are then fed into an LLM to generate the reward function, which in turn is used to train the policy. To enhance the quality of the reward function, we develop a video-assisted iterative reward refinement scheme that visually assesses the learned behaviors and provides textual feedback to the LLM. This feedback guides the LLM to continually refine the reward function, ultimately facilitating more efficient behavior learning. Experimental results on tasks involving bipedal and quadrupedal robot motion control demonstrate that our method surpasses the performance of state-of-the-art LLM-based reward generation methods by over 37.6% in terms of human normalized score. More importantly, by switching video inputs, we find our method can rapidly learn diverse motion behaviors such as walking and running.

Figures

Figures reproduced from arXiv: 2412.05515 by the authors.

Figure 1
Figure 1. Existing large language model-based reward generation meth￾ods [19, 38] rely solely on text as input, which is inefficient due to the diffi￾culty of adequately describing complex behaviors. Our method directly uses videos with target behaviors as inputs, which can be also used to visually as￾sess the learning outcomes of the agent’s behavior, providing feedback to the LLM to facilitate efficient reward generation. c… view at source ↗
Figure 2
Figure 2. Overview of our Video2Reward method. We propose a video-to-text transforming module that converts keypoints of motion in video clips into text. This textual information, alongside auxiliary textual contexts, guides a large language model (LLM) in designing reward functions. After each training cycle, we visually compare the robot’s keypoint trajectories with those of the target in the video, receiving feedback on th… view at source ↗
Figure 3
Figure 3. Comparisons on Anymal and Humanoid tasks in terms of human normalized score and maximum training success. Our method outperforms all baseline methods, including Human, suggesting the introduction of visual information is effective [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Visualization results of quadrupedal robots trained by our method. Specific behaviors are learned effectively from different videos. to capture the coordinates of the object’s joints every three frames, obtaining 2 seconds of video information. This approach ensures th…
Figure 6
Figure 6. Figure 6: Comparisons of the learned running behavior in the Humanoid task. Our running pose is visually the closest to that of a real human. 5.6 Qualitative Results Keypoint trajectory. To validate whether our method facilitates the generation of smoother and more natural movem…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Representative Volume Element: Existence and Extent in Cracked Heterogeneous Medium

    cs.CE 2025-08 unverdicted novelty 5.0 of 10

    Modified periodic boundary conditions that add strain periodicity to displacement periodicity are claimed to reduce mesh and size sensitivity in cracked-composite RVE simulations, tested on 1,200 samples.

Reference graph

Works this paper leans on

40 extracted references · 32 canonical work pages · cited by 1 Pith paper

  1. [1]

    Aractingi, P.-A

    M. Aractingi, P.-A. Léziart, T. Flayols, J. Perez, T. Silander, and P. Souères. Controlling the solo12 quadruped robot with deep rein- forcement learning. scientific Reports, 13(1):11945, 2023

    work page 2023

  2. [2]

    Arora and P

    S. Arora and P. Doshi. A survey of inverse reinforcement learning: Challenges, methods and progress. Artificial Intelligence, 297:103500, 2021

    work page 2021

  3. [3]

    R. D. Beer. A dynamical systems perspective on agent-environment interaction. Artificial intelligence, 72(1-2):173–215, 1995

    work page 1995

  4. [4]

    Bohez, S

    S. Bohez, S. Tunyasuvunakool, P. Brakel, F. Sadeghi, L. Hasenclever, Y . Tassa, E. Parisotto, J. Humplik, T. Haarnoja, R. Hafner, et al. Imitate and repurpose: Learning reusable robot movement skills from human and animal behaviors. arXiv preprint arXiv:2203.17138, 2022

    arXiv 2022

  5. [5]

    Bratko, T

    I. Bratko, T. Urban ˇciˇc, and C. Sammut. Behavioural cloning: phenom- ena, results and problems.IFAC Proceedings Volumes, 28(21):143–149, 1995

    work page 1995

  6. [6]

    Breitenbach, B

    T. Breitenbach, B. Wilkusz, L. Rasbach, and P. Jahnke. On a method for detecting periods and repeating patterns in time series data with au- tocorrelation and function approximation. Pattern Recognition, 138: 109355, 2023

    work page 2023

  7. [7]

    Y . Cai, Z. Wang, Z. Luo, B. Yin, A. Du, H. Wang, X. Zhang, X. Zhou, E. Zhou, and J. Sun. Learning delicate local representations for multi- person pose estimation. In Proceedings of the European Conference on Computer Vision (ECCV), pages 455–472, 2020

    work page 2020

  8. [8]

    Learning predictive representations in autonomous driving to improve deep reinforcement learning

    D. Graves, N. M. Nguyen, K. Hassanzadeh, and J. Jin. Learning predic- tive representations in autonomous driving to improve deep reinforce- ment learning. arXiv preprint arXiv:2006.15110, 2020

    work page Pith review arXiv 2006

Show all 40 references
  1. [9]

    Hadfield-Menell, S

    D. Hadfield-Menell, S. J. Russell, P. Abbeel, and A. Dragan. Cooper- ative inverse reinforcement learning. Advances in neural information processing systems, 29, 2016

    2016

  2. [10]

    Hu and D

    H. Hu and D. Sadigh. Language instructed reinforcement learning for human-ai coordination. In Proceedings of International Conference on Machine Learning (ICML), pages 13584–13598, 2023

    2023

  3. [11]

    Huang, G

    C. Huang, G. Wang, Z. Zhou, R. Zhang, and L. Lin. Reward-adaptive re- inforcement learning: Dynamic policy gradient optimization for bipedal locomotion. IEEE Transactions on Pattern Analysis and Machine Intel- ligence, 45(6):7686–7695, 2023

    2023

  4. [12]

    Hussein, M

    A. Hussein, M. M. Gaber, E. Elyan, and C. Jayne. Imitation learning: A survey of learning methods. ACM Computing Surveys (CSUR), 50(2): 1–35, 2017

    2017

  5. [13]

    S. Kim, M. Sorokin, J. Lee, and S. Ha. Humanconquad: human motion control of quadrupedal robots using deep reinforcement learning. In SIGGRAPH Asia 2022 Emerging Technologies, pages 1–2. 2022

    2022

  6. [14]

    W. Koch, R. Mancuso, R. West, and A. Bestavros. Reinforcement learn- ing for uav attitude control. ACM Transactions on Cyber-Physical Sys- tems, 3(2):1–21, 2019

    2019

  7. [15]

    Kroemer, S

    O. Kroemer, S. Niekum, and G. Konidaris. A review of robot learn- ing for manipulation: Challenges, representations, and algorithms. The Journal of Machine Learning Research, 22(1):1395–1476, 2021

    2021

  8. [16]

    M. Kwon, S. M. Xie, K. Bullard, and D. Sadigh. Reward design with language models. In Proceedings of the International Conference on Learning Representations (ICLR), 2023

    2023

  9. [17]

    Laskey, J

    M. Laskey, J. Lee, R. Fox, A. Dragan, and K. Goldberg. Dart: Noise in- jection for robust imitation learning. In1st Conference on Robot Learn- ing, 2017

    2017

  10. [18]

    J. Lin, D. Fried, D. Klein, and A. Dragan. Inferring rewards from lan- guage in context. arXiv preprint arXiv:2204.02515, 2022

    2022 arXiv

  11. [19]

    Y . J. Ma, W. Liang, G. Wang, D.-A. Huang, O. Bastani, D. Jayaraman, Y . Zhu, L. Fan, and A. Anandkumar. Eureka: Human-level reward de- sign via coding large language models. In Proceedings of the Interna- tional Conference on Learning Representations (ICLR), 2024

    2024

  12. [20]

    Makoviychuk, L

    V . Makoviychuk, L. Wawrzyniak, Y . Guo, M. Lu, K. Storey, M. Mack- lin, D. Hoeller, N. Rudin, A. Allshire, A. Handa, et al. Isaac gym: High performance gpu-based physics simulation for robot learning. arXiv preprint arXiv:2108.10470, 2021

    2021 arXiv

  13. [21]

    Paniego, N

    S. Paniego, N. Paliwal, and J. Cañas. Model optimization in deep learn- ing based robot control for autonomous driving. IEEE Robotics and Automation Letters, 9(1):715–722, 2023

    2023

  14. [22]

    X. B. Peng, E. Coumans, T. Zhang, T. E. Lee, J. Tan, and S. Levine. Learning agile robotic locomotion skills by imitating animals. In Robotics: Science and Systems(RSS)), 2020

    2020

  15. [23]

    Reske, J

    A. Reske, J. Carius, Y . Ma, F. Farshidian, and M. Hutter. Imitation learning from mpc for quadrupedal multi-gait control. In Proceedings of International Conference on Robotics and Automation (ICRA), pages 5014–5020, 2021

    2021

  16. [24]

    Ross and D

    S. Ross and D. Bagnell. Efficient reductions for imitation learning. In Proceedings of the International Conference on Artificial Intelligence and Statistics (IJCAI), pages 661–668, 2010

    2010

  17. [25]

    L. Rozo, P. Jiménez, and C. Torras. A robot learning from demonstra- tion framework to perform force-based manipulation tasks. Intelligent service robotics, 6(1):33–51, 2013

    2013

  18. [26]

    Rudin, H

    N. Rudin, H. Kolvenbach, V . Tsounis, and M. Hutter. Cat-like jumping and landing of legged robots in low gravity using deep reinforcement learning. IEEE Transactions on Robotics, 38(1):317–328, 2021

    2021

  19. [27]

    Salvador and P

    S. Salvador and P. Chan. Toward accurate dynamic time warping in linear time and space. Intelligent Data Analysis, 11(5):561–580, 2007

    2007

  20. [28]

    Schöner, M

    G. Schöner, M. Dose, and C. Engels. Dynamics of behavior: Theory and applications for autonomous robot architectures. Robotics and au- tonomous systems, 16(2-4):213–245, 1995

    1995

  21. [29]

    Schulman, F

    J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov. Prox- imal policy optimization algorithms. arXiv preprint arXiv:1707.06347, Jul 2017

    2017 arXiv

  22. [30]

    Siekmann, S

    J. Siekmann, S. Valluri, J. Dao, L. Bermillo, H. Duan, A. Fern, and J. Hurst. Learning memory-based control for human-scale bipedal lo- comotion. arXiv preprint arXiv:2006.02402, 2020

    2006 arXiv

  23. [31]

    K. Sun, B. Xiao, D. Liu, and J. Wang. Deep high-resolution represen- tation learning for human pose estimation. In Proceedings of the IEEE conference on computer vision and pattern recognition(CVPR) , pages 5693–5703, 2019

    2019

  24. [32]

    R. S. Sutton and A. G. Barto. Reinforcement learning: An introduction. MIT press, 2018

    2018

  25. [33]

    Y . Tang, W. Yu, J. Tan, H. Zen, A. Faust, and T. Harada. Saytap: Language to quadrupedal locomotion. In Conference on Robot Learn- ing(CoRL), volume 229, pages 3556–3570, 2023

    2023

  26. [34]

    Torabi, G

    F. Torabi, G. Warnell, and P. Stone. Behavioral cloning from observa- tion. arXiv preprint arXiv:1805.01954, 2018

    2018 arXiv

  27. [35]

    T. Xie, S. Zhao, C. H. Wu, Y . Liu, Q. Luo, V . Zhong, Y . Yang, and T. Yu. Text2reward: Automated dense reward function generation for reinforcement learning. In Proceedings of the International Conference on Learning Representations (ICLR), 2024

    2024

  28. [36]

    Z. Xie, P. Clary, J. Dao, P. Morais, J. Hurst, and M. Panne. Learning locomotion skills for cassie: Iterative design and sim-to-real. In Con- ference on Robot Learning, pages 317–329, 2020

    2020

  29. [37]

    Y . Yang, K. Caluwaerts, A. Iscen, T. Zhang, J. Tan, and V . Sindhwani. Data efficient reinforcement learning for legged robots. In Conference on Robot Learning (CoRL), pages 1–10, 2020

    2020

  30. [38]

    W. Yu, N. Gileadi, C. Fu, S. Kirmani, K.-H. Lee, M. G. Arenas, H.-T. L. Chiang, T. Erez, L. Hasenclever, J. Humplik, et al. Language to rewards for robotic skill synthesis. arXiv preprint arXiv:2306.08647, 2023

    2023 arXiv

  31. [39]

    J. Yue. Learning locomotion for legged robots based on reinforcement learning: A survey. InInternational Conference on Electrical Engineer- ing and Control Technologies (CEECT), pages 1–7, 2020

    2020

  32. [40]

    Zhang, Z

    K. Zhang, Z. Hou, C. W. de Silva, H. Yu, and C. Fu. Teach biped robots to walk via gait principles and reinforcement learning with adversarial critics. arXiv preprint arXiv:1910.10194, 2019

    1910 arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.