Pith. sign in

REVIEW 3 major objections 4 minor 40 references

SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read SpeedTuning speeds robot policies 2.4x without collecting new demonstrations.

desk verdict SpeedTuning is a clean, practical idea—train a lightweight RL policy to choose execution speeds for a frozen imitation policy—and the paper's experiments are broad enough to warrant serious review, but a technical error in the interpolation equation and a missing promised baseline need fixing first. read the letter →

arxiv 2608.09138 v2 pith:GDVVHIAO submitted 2026-08-10 cs.RO cs.AI

David D. Yuan , Tony Z. Zhao , Kaylee Burns , Chelsea Finn This is my paper · ORCID
classification cs.ROcs.AI
keywords SpeedTuningimitationlearningreinforcementrobotmanipulationactionchunkingtemporalinterpolationpolicyacceleration
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

SpeedTuning claims that an imitation-learned manipulation policy can be run much faster by adding a lightweight, reinforcement-learned speed policy that chooses how fast to execute each chunk of actions. The speed policy is trained on top of a frozen task policy, using a reward that combines task success with a speed incentive, so it learns to slow down for contact-rich or dynamic phases and speed up during less demanding stages. The paper reports over 2.4x speed-ups across six simulated and real tasks while preserving adequate success rates, beating both the original policy and a fixed-speed interpolation baseline.

What carries the argument

The central machinery is the speed policy $\pi_\varphi(v_t|s_t)$ coupled with linear temporal interpolation over action chunks (Eqs. 9-11). The base policy emits a chunk of joint-position actions; the speed policy selects a discrete speed multiplier; the chunk is re-sampled in time by interpolation, effectively shrinking the chunk length to $(k+1)/v$. This single scalar re-tempos the frozen task policy, and the reinforcement-learning objective $r_{ST} = \alpha v^{\beta} + r_{task}$ gives the speed policy a task-sensitive speed schedule.

What would settle it

Run the same base policy on a contact-rich task at the episode-average speed SpeedTuning chooses, replacing the speed policy with a fixed multiplier of that value; if success matches SpeedTuning, the claim that state-dependent speed selection is necessary would fail. More directly, compare the rollout from executing an interpolated action chunk at 2x against the original chunk at 1x under identical dynamics; if contact forces or control frequency cause the 2x rollout to diverge from the underlying planned trajectory, the time-scaling premise fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that execution speed can be treated as a controllable action dimension on top of a frozen imitation policy, without retraining the task policy or collecting new demonstrations. SpeedTuning factorizes the policy into a task policy and a speed policy, and uses linear temporal interpolation over action chunks to convert the speed policy's chosen multiplier into accelerated joint-position commands. The speed policy observes RGB images, joint positions, and joint speeds, outputs a multiplier from a discrete set, and is optimized with Rainbow DQN; it deliberately lowers speed at critical moments such as grasping, pouring, or synchronizing with a swinging tea bag, and raises it during less critical phases. Across cube transfer, peg insertion, tea bag transfer, tea bag disposal, food preparation, and almond pouring, the paper reports over 2.4x speed-up with success comparable to the base policy and better than uniform acceleration.

Load-bearing premise

The base policy's action chunks can be faithfully accelerated by linear temporal interpolation, so executing the time-warped action sequence at speed $v$ actually produces the same task-relevant trajectory in $1/v$ of the time.

Editorial extensions

If this is right

  • Existing imitation policies can be accelerated at deployment time without recollecting demonstrations or retraining the base policy.
  • State-dependent speed selection improves the speed-success Pareto frontier, especially on dynamic tasks where fixed multipliers fail.
  • The method applies across different chunked task-policy architectures and transfers from simulation to real-world manipulation.
  • The learned speed policy effectively segments a task into critical and non-critical phases, yielding interpretable slow-down and speed-up behavior.

Reading between the lines

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

  • Inference: The same speed-policy interface could accelerate other chunked visuomotor policies, such as diffusion-based policies, without retraining, making SpeedTuning a general post-hoc accelerator for imitation-learned manipulation.
  • Inference: Co-training the speed policy jointly with the task policy might recover success in contact-rich settings where pure interpolation distorts the planned trajectory, because the task policy could adapt its actions to the selected tempo.
  • Inference: The speed policy's reliance on image observations suggests it is learning phase detection; a cheaper task-progress estimator could potentially replace the RL-trained component and yield similar speed-ups, separating 'when to go fast' from 'how to go fast'.
  • Inference: The $\alpha$ and $\beta$ reward parameters define a tunable speed-success frontier, so a practical deployment could automatically select these weights to meet a target success-rate constraint.
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

3 major / 4 minor

Summary. The paper introduces SpeedTuning, a reinforcement learning framework that trains a discrete-action speed policy on top of a frozen imitation learning policy. The speed policy outputs a speed multiplier, and action chunks from the base policy are linearly interpolated in time according to that multiplier. The combined reward is a weighted sum of a speed reward and the task success reward, and the speed policy is trained with Rainbow DQN. The authors evaluate the method on three simulated and three real-world ALOHA manipulation tasks, reporting over 2.4x speed-ups with preserved success rates relative to the original policy and to a fixed-speed interpolation baseline, and they provide ablation studies over the task policy, image observations, speed reward exponent, and frame skip.

Significance. If the central claim holds, SpeedTuning offers a practical way to retrofit existing imitation-learned policies with faster execution without collecting new demonstrations, and the paper would make a useful contribution to the growing literature on efficient robot manipulation. The work has several concrete strengths: the method is clearly described, the RL formulation is standard, code and videos are promised, and the ablation studies cover important design choices. The reported result that a learned adaptive speed policy can outperform a fixed-speed baseline on dynamic and precise tasks is interesting and plausible. However, the correctness of the reported speed-ups depends on a precise and correct mapping between the speed multiplier and actual time scaling, and this part of the manuscript currently contains a technical error that needs to be resolved before the claims can be accepted.

major comments (3)
  1. [Section III-F, Eq. (9)] Equation (9) does not implement linear temporal interpolation as claimed. Standard time scaling of a function f at speed v evaluates f(vt), with fractional interpolation weight vt - floor(vt) between f(floor(vt)) and f(floor(vt)+1). The extra division by v in Eq. (9) gives weight (vt - floor(vt))/v instead. For f(i)=i, v=1.5, and i=1, the correctly time-scaled value is 1.5, whereas Eq. (9) yields 1 + 0.5/1.5 = 1.333. Consequently, the multiplier v in Eqs. (10)-(11) does not correspond to uniform time scaling of the base action chunk. Moreover, the number of executed steps in Eq. (11) is floor((k+1)/v)+1, which is larger than (k+1)/v for non-integer v, so the effective speed-up is smaller than v. Because the speed reward and the fixed-speed baseline sweep in Fig. 4 both rely on interpreting v as an actual speed multiplier, this issue is load-bearing for the central claim. Please correct the interpolation formula or explicitly define the non-uniform mapping actually used, and re-derive the effective speed-up and the Pareto plots accordingly.
  2. [Section IV-C and Fig. 4] The headline claim of "over 2.4x speed-up" is never precisely defined. It is unclear whether speed-up is measured as total episode wall-clock time, total number of environment steps, average selected speed multiplier, or the ratio of original episode length to accelerated episode length. Equations (9)-(11) suggest a step-count reduction per chunk, but the episodic speed-up also depends on how non-integer speeds and the final partial chunk are handled. Please state the exact formula used to compute the reported speed-ups and clarify whether results are based on environment steps or physical time. Without this definition, the reader cannot verify the central quantitative claim.
  3. [Section III-F and Section IV-C] The method's core assumption is that linearly interpolating the joint-position action chunk at speed v faithfully produces the same underlying task trajectory in 1/v of the time. This assumption is not verified in the paper, and it is especially nontrivial for contact-rich and dynamic tasks where physical dynamics, friction, and control frequency can make time-scaled action sequences invalid. Even after correcting Eq. (9), the authors should provide direct evidence for this assumption, for example by comparing the interpolated action execution against executing the original action sequence at a higher hardware command rate, or by measuring the actual trajectory achieved under both schemes. If the assumption fails, the speed policy cannot fully compensate, since it only selects multipliers over interpolated actions.
minor comments (4)
  1. [Section I vs. Section IV] The introduction states that the method is compared "against reinforcement learning from scratch and a naive speed-up baseline," but the experiments only report the fixed-speed interpolation baseline. Either add the RL-from-scratch comparison or revise the introduction to remove the unfulfilled promise.
  2. [Section I and Section III] The phrase "requires no extra data-collection" is misleading: training the speed policy with reinforcement learning requires collecting environment interaction samples. The intended meaning is presumably "no additional human demonstrations." Please clarify this in the text.
  3. [Fig. 4 caption] The caption notes that the real-world speed policy was not trained to full convergence. This is an important limitation that should appear in the main text as well, and the abstract's claim of demonstrating "robustness in real-world robotic manipulation" should be moderated accordingly.
  4. [Section V and Fig. 6] The main experiments use beta=2 and frame skip=10, which are selected via ablations on Simulated Tea Bag Transfer. Please state explicitly whether the same hyperparameters were used across all six tasks, and report the sensitivity of the main results to this choice.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the speed-ups and Pareto-frontier comparisons in SpeedTuning are empirically measured outcomes of RL optimization, not definitional consequences of its reward or of cited prior work.

full rationale

I walked the paper's derivation chain from the imitation-learning objective (Eq. 1) through the SpeedTuning reward (Eqs. 3, 5, 7) to the interpolation mechanism (Eqs. 9-11) and the experimental claims. The speed policy is trained by RL to maximize a combined reward that includes both speed and task success, and the speed-policy class does contain constant-speed policies as a special case; however, the paper does not derive its 2.4x speed-up or its success rates from this containment. The reported numbers come from actual rollouts, and the RL optimization could in principle fail or generalize poorly, so the comparison to the fixed-speed baseline is an empirical benchmark rather than a fitted prediction. The cited ACT base policy [1] is prior work used as a component, not a self-citation carrying the central claim, and there is no imported uniqueness theorem or ansatz hidden in a citation. The interpolation equation (Eq. 9) may contain a technical inconsistency in how non-integer speeds are implemented, but that is a correctness concern, not circularity. Likewise, selecting beta and frame skip via ablations on Simulated Tea Bag Transfer and then evaluating on that same task is a potential overfitting/selection concern, but it does not make the method's outputs equivalent to its inputs. No fitted parameter is renamed as a prediction, and no known result is repackaged as new. I therefore find no significant circularity.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central claim depends on hyperparameters (alpha, beta, k_skip, V) that are either selected via ablations on the main evaluation task or not reported. The method also assumes binary success rewards and valid temporal time-scaling of action chunks. No new physical entities are introduced.

free parameters (6)
  • alpha (speed reward weight) = not reported
    Weighting term in Eq 3 and Eq 7 that trades off speed reward against task reward; no value is given in the text, and the central speed-success results depend on it.
  • beta (speed reward exponent) = 2
    Exponent in r_speed(v) = v^beta; chosen from ablations in Appendix D on Simulated Tea Bag Transfer, which is also a main evaluation task.
  • frame skip k_skip = 10
    Interval between speed-policy decisions; selected from ablations in Appendix E, affecting control granularity and MDP horizon.
  • discrete speed set V = not reported
    The set of allowed speed multipliers used by the speed policy; the paper mentions a discrete set but does not list the values.
  • discount factor gamma = not reported
    Standard RL discount factor in Eq 8; not specified.
  • frame stack k_stack = not reported
    Number of recent observations fed to the speed policy; mentioned in Section III-E but not quantified.
assumptions (5)
  • domain assumption The binary task reward r_task(s_t, a_t) = 1{s_{t+1} in S_success} (Eq 2) adequately captures task success.
    The entire RL reward signal in Eq 7 assumes success is a sparse binary event; in real-world tasks the paper says this can be provided by human feedback, but the labeling procedure is not detailed.
  • domain assumption Linear temporal interpolation (Eq 9) faithfully time-scales action trajectories.
    The method's core mechanism assumes that interpolating action chunks at speed v produces a valid faster execution of the original plan; this may break under contact, friction, or control-bandwidth limits.
  • domain assumption The observation history (frame stack of proprioception and images) contains enough information for the speed policy to identify when to speed up or slow down.
    The speed policy is a function of the observation stack; if critical task phases are not visually or kinesthetically distinguishable, the policy cannot adapt correctly.
  • standard math The Bellman equation for discrete Q-learning (Eq 8) is a valid optimization target for the speed MDP.
    Rainbow DQN is a standard value-based RL algorithm; the MDP is assumed to be Markovian given the frame stack.
  • ad hoc to paper The additive reward structure r_ST = alpha*r_speed(v) + r_task (Eq 7) is an appropriate scalarization of the two objectives.
    The linear weighting of speed and success is a design choice, not a derived consequence; different scalarizations could yield different Pareto points.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning." pith.science (2026). https://pith.science/paper/GDVVHIAO

@misc{pith2026260809138,
  author       = {Pith},
  title        = {Pith review of: SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GDVVHIAO}},
  note         = {Machine review of arXiv:2608.09138}
}
read the original abstract

While learned robotic policies hold promise for advancing generalizable manipulation, their practical deployment is often hindered by suboptimal execution speeds. Imitation learning policies are inherently limited by hardware constraints and the speed of the operator during data collection. In addition, there are no established methods for accelerating policies learned via imitation, and the empirical relationship between execution speed and task success remains underexplored. To address these issues, we introduce SpeedTuning, a reinforcement learning framework specifically designed to enhance the speed of manipulation policies. SpeedTuning learns to predict the optimal execution speed for actions, thereby complementing a base policy without necessitating additional data collection. We provide empirical evidence that SpeedTuning achieves substantial improvements in execution speed, exceeding 2.4x speed-up, while preserving an adequate success rate compared to both the original task policy and straightforward speed-up methods such as linear interpolation at a fixed speed. We evaluate our approach across a diverse set of dynamic and precise tasks, including pouring, throwing, and picking, demonstrating its effectiveness and robustness in enhancing real-world robotic manipulation. Videos and code are available at https://daivdyuan.github.io/speed-tuning/

Figures

Figures reproduced from arXiv: 2608.09138 by the authors.

Figure 1
Figure 1. SPEEDTUNING is a reinforcement learning framework designed to optimize both execution speed and task success for learned manipulation policies. The left panel presents a schematic of SPEEDTUNING augmenting an imitation learning policy, where it outputs a speed multiplier to modulate the execution of predicted actions. This speed policy is optimized using reinforcement learning. The right panel illustrates performanc… view at source ↗
Figure 3
Figure 3. Initialization and key actions for three real-world tasks (top to bottom): Tea Bag Disposal, Food Preparation, and Almond Pouring. an additional hyperparameter, β, to further scale the speed reward: rspeed(v) = v β . To manage complexity in long￾horizon tasks, we use a frame skip constant kskip ∈ Z +, querying the speed policy only once every kskip steps, while continuing executing the accelerated policy with the la… view at source ↗
Figure 4
Figure 4. Results showing the SPEEDTUNING policy on simulated (top row) and real world (bottom row) tasks. For simulated tasks, the baseline curve is obtained by evaluating 100 episodes across various speeds at 0.1 intervals, while the distribution and mean of the SPEEDTUNING policy is calculated over 2000 episodes. For the real-world tasks, the baseline curve is obtained by evaluating 20 episodes across various speeds at 0.2… view at source ↗
Figures from the paper (8 more)
Figure 5
Figure 5. Figure 5: Comparison of the SPEEDTUNING and vanilla ACT policy trajectories for the Almond Pouring task, featuring key action captures and speed-up values throughout the task. The results highlight SPEEDTUNING’s task understanding, significantly accelerating overall execution by…
Figure 6
Figure 6. Figure 6: Ablation study on task policy, speed rewards, and frame skip values. All configurations are evaluated over 2000 episodes on Simulated Tea Bag Transfer. bimanual and dynamic manipulation. The robot must first pick up a mug containing a tea bag and then carefully dispose…
Figure 7
Figure 7. Figure 7: provides a qualitative view of SPEEDTUNING on the three simulated tasks. Cube Transfer proceeds from grasping to a bimanual handoff; Peg Insertion requires grasping, alignment, and insertion; and Tea Bag Transfer requires transporting the suspended tea bag into the rec…
Figure 8
Figure 8. Figure 8: Base-policy ablation: acceleration (left) and success rate (right). Learned ACT is compared with an open-loop scripted policy. [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]
Figure 9
Figure 9. Figure 9: Visual-observation ablation: acceleration (left) and success rate (right). Both variants retain proprioception. [PITH_FULL_IMAGE:figures/full_fig_p009_9.png]
Figure 10
Figure 10. Figure 10: Speed-reward ablation: acceleration (left) and success rate (right). The main experiments use [PITH_FULL_IMAGE:figures/full_fig_p010_10.png]
Figure 11
Figure 11. Figure 11: Frame-skip ablation: acceleration (left) and success rate (right). The main experiments use [PITH_FULL_IMAGE:figures/full_fig_p010_11.png]
Figure 12
Figure 12. Figure 12: Image-encoder ablation: acceleration (left) and success rate (right). [PITH_FULL_IMAGE:figures/full_fig_p010_12.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 8 linked inside Pith

  1. [1]

    Learn- ing fine-grained bimanual manipulation with low-cost hardware,

    T. Z. Zhao, V . Kumar, S. Levine, and C. Finn, “Learn- ing fine-grained bimanual manipulation with low-cost hardware,”arXiv preprint arXiv:2304.13705, 2023

    arXiv 2023

  2. [2]

    Alvinn: An autonomous land ve- hicle in a neural network,

    D. A. Pomerleau, “Alvinn: An autonomous land ve- hicle in a neural network,” inNIPS, 1988

    1988

  3. [3]

    Is imitation learning the route to humanoid robots?

    S. Schaal, “Is imitation learning the route to humanoid robots?”Trends in Cognitive Sciences, vol. 3, pp. 233– 242, 1999. [Online]. Available:https : / / api . semanticscholar.org/CorpusID:7124120

    1999

  4. [4]

    The dlr bimanual haptic device with optimized workspace,

    T. Hulin et al., “The dlr bimanual haptic device with optimized workspace,”2011 IEEE International Con- ference on Robotics and Automation, pp. 3441–3442, 2011

    2011

  5. [5]

    A low cost modular actuator for dynamic robots,

    B. Katz, “A low cost modular actuator for dynamic robots,” 2018

    2018

  6. [6]

    Nimbro avatar: Interactive immersive telepresence with force-feedback telemanipulation,

    M. Schwarz, C. Lenz, A. Rochow, M. Schreiber, and S. Behnke, “Nimbro avatar: Interactive immersive telepresence with force-feedback telemanipulation,” 2021 IEEE/RSJ International Conference on Intel- ligent Robots and Systems (IROS), pp. 5312–5319, 2021

    2021

  7. [7]

    P. Wu, Y . Shentu, Z. Yi, X. Lin, and P. Abbeel, Gello: A general, low-cost, and intuitive teleoperation framework for robot manipulators, 2023

    2023

  8. [8]

    Attention is all you need,

    A. Vaswani et al., “Attention is all you need,”ArXiv, vol. abs/1706.03762, 2017

    arXiv 2017

Show all 40 references
  1. [9]

    Diffusion policy: Visuomotor pol- icy learning via action diffusion,

    C. Chi et al., “Diffusion policy: Visuomotor pol- icy learning via action diffusion,”arXiv preprint arXiv:2303.04137, 2023

    2023 arXiv

  2. [10]

    Implicit behavioral cloning,

    P. R. Florence et al., “Implicit behavioral cloning,” ArXiv, vol. abs/2109.00137, 2021

    2021 arXiv

  3. [11]

    Bc-z: Zero-shot task generalization with robotic imitation learning,

    E. Jang et al., “Bc-z: Zero-shot task generalization with robotic imitation learning,” inConference on Robot Learning, 2022

    2022

  4. [12]

    Should i run offline reinforcement learning or behavioral cloning?

    A. Kumar, J. Hong, A. Singh, and S. Levine, “Should i run offline reinforcement learning or behavioral cloning?” InInternational Conference on Learning Representations, 2022. [Online]. Available:https: //openreview.net/forum?id=AP1MKT37rJ

    2022

  5. [13]

    Q-learning,

    C. Watkins and P. Dayan, “Q-learning,”Machine Learning, vol. 8, pp. 279–292, 1992

    1992

  6. [14]

    R. S. Sutton and A. G. Barto,Reinforcement learning - an introduction(Adaptive computation and machine learning). MIT Press, 1998,ISBN: 978-0-262-19398-6

    1998

  7. [15]

    Policy gradient methods for reinforcement learning with function approximation,

    R. S. Sutton, D. A. McAllester, S. Singh, and Y . Mansour, “Policy gradient methods for reinforcement learning with function approximation,” inNeural In- formation Processing Systems, 1999

    1999

  8. [16]

    Actor-critic algo- rithms,

    V . R. Konda and J. N. Tsitsiklis, “Actor-critic algo- rithms,” inNeural Information Processing Systems, 1999

    1999

  9. [17]

    Continuous control with deep reinforcement learning,

    T. P. Lillicrap et al., “Continuous control with deep reinforcement learning,”CoRR, vol. abs/1509.02971, 2015

    2015 arXiv

  10. [18]

    Deep reinforcement learning for robotic manipulation with asynchronous off-policy updates,

    S. S. Gu, E. Holly, T. P. Lillicrap, and S. Levine, “Deep reinforcement learning for robotic manipulation with asynchronous off-policy updates,”2017 IEEE International Conference on Robotics and Automation (ICRA), pp. 3389–3396, 2016

    2017

  11. [19]

    Learning dexterous in-hand manipulation,

    M. Andrychowicz et al., “Learning dexterous in-hand manipulation,”The International Journal of Robotics Research, vol. 39, pp. 20–3, 2018

    2018

  12. [20]

    How to train your robot with deep reinforcement learning: Lessons we have learned,

    J. Ibarz, J. Tan, C. Finn, M. Kalakrishnan, P. Pastor, and S. Levine, “How to train your robot with deep reinforcement learning: Lessons we have learned,”The International Journal of Robotics Research, vol. 40, pp. 698–721, 2021

    2021

  13. [21]

    Never stop learning: The effectiveness of fine-tuning in robotic reinforcement learning,

    R. C. Julian, B. Swanson, G. S. Sukhatme, S. Levine, C. Finn, and K. Hausman, “Never stop learning: The effectiveness of fine-tuning in robotic reinforcement learning,” inConference on Robot Learning, 2020

    2020

  14. [22]

    Robot fine-tuning made easy: Pre-training rewards and policies for autonomous real-world rein- forcement learning,

    J. Yang, M. S. Mark, B. Vu, A. Sharma, J. Bohg, and C. Finn, “Robot fine-tuning made easy: Pre-training rewards and policies for autonomous real-world rein- forcement learning,”2024 IEEE International Confer- ence on Robotics and Automation (ICRA), pp. 4804– 4811, 2023

    2024

  15. [23]

    Learning to play table tennis from scratch using muscular robots,

    D. B ¨uchler, S. Guist, R. Calandra, V . Berenz, B. Scholkopf, and J. Peters, “Learning to play table tennis from scratch using muscular robots,”IEEE Transactions on Robotics, vol. 38, pp. 3850–3860, 2020

    2020

  16. [24]

    Achieving human level competitive robot table tennis,

    D. B. D’Ambrosio et al., “Achieving human level competitive robot table tennis,” 2024

    2024

  17. [25]

    Knotting manipulation of a flexible rope using a high-speed multifingered hand and high-speed visual and tactile sensory feedback,

    Y . Yamakawa, A. Namiki, M. Ishikawa, and M. Shi- mojo, “Knotting manipulation of a flexible rope using a high-speed multifingered hand and high-speed visual and tactile sensory feedback,”Journal of the Robotics Society of Japan, vol. 27, pp. 1016–1024, 2009

    2009

  18. [26]

    Motion planning for dynamic knotting of a flexible rope with a high-speed robot arm,

    Y . Yamakawa, A. Namiki, and M. Ishikawa, “Motion planning for dynamic knotting of a flexible rope with a high-speed robot arm,”2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 49– 54, 2010

    2010

  19. [27]

    Dynamic high-speed knotting of a rope by a manipulator,

    Y . Yamakawa, A. Namiki, and M. Ishikawa, “Dynamic high-speed knotting of a rope by a manipulator,” International Journal of Advanced Robotic Systems, vol. 10, 2013

    2013

  20. [28]

    Robots of the lost arc: Learning to dynam- ically manipulate fixed-endpoint ropes and cables,

    H. Zhang, J. Ichnowski, D. Seita, J. Wang, and K. Goldberg, “Robots of the lost arc: Learning to dynam- ically manipulate fixed-endpoint ropes and cables,” ArXiv, vol. abs/2011.04840, 2020

    2011 arXiv

  21. [29]

    Tossingbot: Learning to throw arbitrary objects with residual physics,

    A. Zeng, S. Song, J. Lee, A. Rodriguez, and T. Funkhouser, “Tossingbot: Learning to throw arbitrary objects with residual physics,”Ieee Transactions On Robotics, 2019.DOI:10 . 1109 / TRO . 2020 . 2988642

    2019

  22. [30]

    Flingbot: The unreasonable effec- tiveness of dynamic manipulation for cloth unfolding,

    H. Ha and S. Song, “Flingbot: The unreasonable effec- tiveness of dynamic manipulation for cloth unfolding,” inConference on Robot Learning, 2021

    2021

  23. [31]

    Par4: Very high speed parallel robot for pick-and-place,

    V . Nabat, M. de la O Rodr ´ıguez, O. Company, S. Krut, and V . Pierrot, “Par4: Very high speed parallel robot for pick-and-place,”2005 IEEE/RSJ Interna- tional Conference on Intelligent Robots and Systems, pp. 553–558, 2005

    2005

  24. [32]

    Toward fast and optimal robotic pick-and-place on a moving con- veyor,

    S. D. Han, S. W. Feng, and J. Yu, “Toward fast and optimal robotic pick-and-place on a moving con- veyor,”IEEE Robotics and Automation Letters, vol. 5, pp. 446–453, 2019

    2019

  25. [33]

    Critically fast pick-and- place with suction cups,

    H. Pham and Q.-C. Pham, “Critically fast pick-and- place with suction cups,”2019 International Confer- ence on Robotics and Automation (ICRA), pp. 3045– 3051, 2018

    2019

  26. [34]

    Rainbow: Combining improvements in deep reinforcement learning,

    M. Hessel et al., “Rainbow: Combining improvements in deep reinforcement learning,” inProceedings of the AAAI conference on artificial intelligence, vol. 32, 2018

    2018

  27. [35]

    Auto-encoding varia- tional bayes,

    D. P. Kingma and M. Welling, “Auto-encoding varia- tional bayes,”arXiv preprint arXiv:1312.6114, 2013

    2013 arXiv

  28. [36]

    Learning structured output representation using deep conditional genera- tive models,

    K. Sohn, H. Lee, and X. Yan, “Learning structured output representation using deep conditional genera- tive models,”Advances in neural information process- ing systems, vol. 28, 2015

    2015

  29. [37]

    Deep reinforce- ment learning with double q-learning,

    H. v. Hasselt, A. Guez, and D. Silver, “Deep reinforce- ment learning with double q-learning,” inProceedings of the Thirtieth AAAI Conference on Artificial Intelli- gence, ser. AAAI’16, Phoenix, Arizona: AAAI Press, 2016, pp. 2094–2100

    2016

  30. [38]

    Prioritized experience replay,

    T. Schaul, J. Quan, I. Antonoglou, and D. Silver, “Prioritized experience replay,”CoRR, vol. abs/1511.05952, 2015. [Online]. Available: https : / / api . semanticscholar . org / CorpusID:13022595

    2015 arXiv

  31. [39]

    Dueling network architectures for deep reinforcement learning,

    Z. Wang, T. Schaul, M. Hessel, H. Van Hasselt, M. Lanctot, and N. De Freitas, “Dueling network architectures for deep reinforcement learning,” inPro- ceedings of the 33rd International Conference on In- ternational Conference on Machine Learning - Volume 48, ser. ICML’16, New ...

    2016

  32. [40]

    A dis- tributional perspective on reinforcement learning,

    M. G. Bellemare, W. Dabney, and R. Munos, “A dis- tributional perspective on reinforcement learning,” in Proceedings of the 34th International Conference on Machine Learning - Volume 70, ser. ICML’17, Sydney, NSW, Australia: JMLR.org, 2017, pp. 449–458. APPENDIX SIMULATIONROLL...

    2017

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.