Pith. sign in

REVIEW 4 major objections 3 minor 2 cited by

Constrained Diffusers for Safe Planning and Control

T0 review · 4 major / 3 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The paper shows that safety constraints can be enforced on pretrained diffusion planners by replacing the reverse denoising updates with constrained Langevin sampling steps—projected, primal-dual, and augmented Lagrangian—with no…

desk verdict A promising plug-and-play safety idea for diffusion planners is undermined by a mis-derived Langevin step and a missing proof, but the experiments and DCBF integration make it worth reviewing. read the letter →

arxiv 2506.12544 v1 pith:JFAOXD6O submitted 2025-06-14 eess.SY cs.ROcs.SY

Jichen Zhang , Liqun Zhao , Antonis Papachristodoulou , Jack Umenberger This is my paper · ORCID
classification eess.SYcs.ROcs.SY
keywords diffusionmodelsconstrainedsamplingLangevindynamicscontrolbarrierfunctionssafeplanningtrajectoryoptimizationinversesafety-critical
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 is trying to establish that a pretrained diffusion model for trajectory planning can be made constraint-satisfying at deployment time, with no retraining and no change to the network architecture. The central move is to reinterpret the reverse denoising loop as stochastic gradient Langevin sampling and to insert three constrained-optimization mechanisms—projection, primal-dual updates, and an augmented Lagrangian—so that every denoising step steers the trajectory toward the safe set while still following the learned data distribution. Discrete control barrier functions express safety requirements on the generated trajectory, and an inverse dynamics model converts the safe states into actions for closed-loop control. If the claim holds, safety becomes a plug-in property of the sampler rather than something that must be learned or solved as a separate quadratic program at each step.

What carries the argument

The load-bearing identity is the reformulation of the DDPM reverse update as Langevin sampling: $x_{t-1} = x_t + \frac{\beta_t}{2}\nabla_{x_t}\log p(x_t) + \sqrt{\beta_t}z$, where the score is recovered from the denoiser as $\nabla_{x_t}\log p(x_t) = -\epsilon_\theta(x_t,t)/\sqrt{1-\bar\alpha_t}$. This identity turns the pretrained denoiser into a score oracle, onto which the paper grafts three constraint mechanisms: a projection operator $\Pi_C$ onto the feasible set; a primal-dual step that carries a Lagrange multiplier $\lambda$ for $\mathbb{E}_{x\sim q}[g(x)] \le 0$; and an augmented-Lagrangian step with slack variable $s$ and penalty $\rho$. Safety at the trajectory level is expressed by the discrete control barrier function inequality $h(x_{\tau+1}) \ge (1-\alpha)h(x_\tau)$, and an inverse dynamics model $u_\tau = \mathrm{IDM}(x_\tau, x_{\tau+1})$ maps the resulting safe state transitions to control actions.

What would settle it

Run the primal-dual or augmented-Lagrangian sampler on a two-dimensional Gaussian with a known score and a nonconvex constraint such as the outside of a disk, and measure the fraction of final samples that violate the constraint as the diffusion horizon increases; if the violation fraction does not approach zero while the samples stay close to the unconstrained target, the claimed equivalence to constrained Langevin sampling is false.

Watch

Extended reading notes

Core claim

The central claim is that constrained planning can be solved entirely inside the reverse diffusion process. The paper rewrites the standard DDPM reverse update as a Langevin sampling step whose score is supplied by the neural denoiser, so that adding safety constraints reduces to modifying each denoising update: project onto the constraint set, subtract the constraint gradient weighted by a Lagrange multiplier, or apply an augmented-Lagrangian penalty with a slack variable. Theorem 4.1 states that the primal-dual version converges almost surely to a local saddle point—equivalently a local optimum of the minimum-KL constrained problem—as the diffusion horizon $T$ tends to infinity. The experiments in Maze2D, locomotion, and a moving-obstacle ball-running task are offered as evidence that the finite-horizon versions keep constraint violations near zero at computation times close to the unconstrained baseline and far below the CBF-QP baseline.

Load-bearing premise

The load-bearing premise is that the DDPM reverse update is exactly a Langevin sampling step with the learned score and the same step size, despite dropping the $1/\sqrt{1-\beta_t}$ rescaling; if that step-size or score approximation is materially wrong, the safety and convergence claims do not follow.

Editorial extensions

If this is right

  • A pretrained trajectory diffuser can serve multiple safety specifications at deployment; only the sampling loop changes.
  • Constraint satisfaction no longer requires solving a quadratic program at every diffusion step, so safe planning runs at near-unconstrained speed.
  • Time-varying constraints, such as moving obstacles, can be handled online by recomputing the DCBF condition from the latest observation and resampling.
  • In the infinite-horizon limit with the correct score, the primal-dual updates converge almost surely to a local saddle point of the constrained sampling problem.
  • Closed-loop control inherits safety only when the inverse dynamics model accurately maps safe state transitions to actions.

Reading between the lines

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

  • Editorial inference: the same constrained-Langevin surgery could be applied to any differentiable constraint on any diffusion generative model, not just trajectory planners; image and video editors could enforce geometric or semantic constraints without retraining.
  • Editorial inference: Theorem 4.1 is asymptotic in the diffusion horizon and assumes the score is the true score; finite-horizon runs with learned scores need empirical safety margins because the theorem does not quantify the gap.
  • Editorial inference: the method assumes an inverse dynamics model exists and that inputs are unconstrained; systems with actuation limits or underactuation would require folding input constraints into the DCBF condition or adding a feasible-input projection.
  • Editorial inference: because safety is enforced only during generation, a miss by the learned score or the constraint gradient is uncaught; layering a lightweight control-barrier filter on the executed action would bound worst-case violations.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 3 minor

Summary. The paper proposes Constrained Diffusers, a framework for enforcing safety and trajectory constraints on pre-trained diffusion planning models without retraining. It rewrites the denoising update as a Langevin sampling step and then applies projected, primal-dual, and augmented-Lagrangian updates (Eqs. (10), (13), (16)) to the reverse diffusion process. Discrete control barrier functions and an inverse dynamics model are introduced for closed-loop safety. Experiments on Maze2D, MuJoCo locomotion, and PyBullet ball running compare the three variants against Diffuser, Conditional Diffuser, and SafeDiffuser. Theorem 4.1 claims almost-sure convergence to a local saddle point / local optimum of the constrained sampling problem (9) as T goes to infinity.

Significance. If the derivations were correct, the framework would be a practically attractive plug-and-play route to constraint satisfaction without retraining: the three algorithms are simple to implement, the DCBF formulation matches trajectory-level safety requirements, and the empirical results show near-zero violations with lower computation time than SafeDiffuser. The paper also deserves credit for evaluating both static and time-varying constraints across several benchmarks. However, the validity of the central 'Langevin reformulation' and of the convergence theorem is currently not established; the missing appendices and the algebraic gap in Section 2.1 mean that the paper's theoretical value cannot be assessed as submitted.

major comments (4)
  1. [Section 2.1, Eqs. (4)-(6)] The claimed reformulation of the DDPM update (4) as the Langevin update (6) is algebraically incorrect. Substituting the Tweedie score relation (5) into (4) gives x_{t-1} = (1/sqrt(1-beta_t))(x_t + beta_t grad_{x_t} log p_t(x_t)) + sqrt(beta_t) z, not Eq. (6). Expanding 1/sqrt(1-beta_t) to first order adds a drift (beta_t/2)x_t and leaves the score coefficient as beta_t rather than beta_t/2, so Eq. (6) does not describe the reverse process of the pretrained DDPM. Because Eqs. (10), (13) and (16) all use Eq. (6), the constrained sampling procedure is not sampling from the distribution p(x_0:T) defined by the original diffusion model, and the KL objective in (9) is not the objective actually being optimized. This gap also prevents transferring the convergence guarantees of [12-14] to the proposed updates.
  2. [Section 4.2, Theorem 4.1 and Section 4.1/4.3] Theorem 4.1 is the central theoretical claim of the paper, but its proof is deferred to Appendix E, and Appendices D and F, which are supposed to contain the analyses for the projected and ALM methods, are not present in the manuscript. The 'mild conditions' of the theorem are not stated, and the object whose convergence is claimed (the sequence of measures induced by the sample path, or the sample path itself?) is never defined. As submitted, the theorem is an assertion rather than a verifiable result, and the same problem affects the convergence statements in Sections 4.1 and 4.3.
  3. [Section 4.2, Eqs. (13)-(14)] The dual update in Eq. (14) and the slack/dual updates in Eqs. (17)-(18) require the expectation E_{x~q}[g(x_t)] over the current distribution, but the algorithm maintains only a single trajectory. If this expectation is approximated by the instantaneous value g(x_t), the resulting stochastic approximation has a bias that must be analyzed; if a batch-based estimator is intended, it is not specified. Without an explicit oracle for these expectations, the 'almost surely' convergence statement of Theorem 4.1 cannot be evaluated, and the pseudocode in Algorithm 1 is incomplete.
  4. [Section 4.2, Theorem 4.1 and Section 2.1] The reverse diffusion process is finite-horizon and time-inhomogeneous: the score p_t changes with t and the index descends from T to 0. The Langevin convergence results in [12-14] apply to stationary, time-homogeneous dynamics with step sizes tending to zero. The phrase 'as T goes to infinity' does not by itself bridge this gap because no rescaling or time-change of the diffusion schedule is provided. Even if the missing appendix were supplied, Theorem 4.1 would need a separate argument for the time-inhomogeneous finite-horizon setting.
minor comments (3)
  1. [Algorithm 1, line 9] The line `u_tau = IDM(x0_0, x1_0)` uses fixed indices 0 and 1 rather than the environment time index tau; presumably x_tau^0 and x_{tau+1}^0 are intended. Please correct.
  2. [Section 6, Table 2] In Table 2, Projected has a per-step time of 0.9305 s for Hopper and 0.9049 s for Swimmer, which is roughly 300 times larger than Primal-Dual and ALM, yet the discussion claims computational efficiency for the proposed methods; the cause of this discrepancy (presumably the cost of projection) should be discussed explicitly.
  3. [Section 4, Eqs. (13)-(14)] The index convention in the updates is confusing: the iteration descends from T to 0, but the primal update (13) and dual update (14) both write t-1 on the left while using lambda_t on the right. The direction of iteration and the role of lambda_t should be clarified to make the algorithm unambiguous.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the constrained-sampling convergence results are imported from external prior work, and the experimental metrics are not fitted inputs.

full rationale

I walked the paper's claimed derivation chain from the reverse diffusion update (Eq. (4)) through the constrained sampling algorithms in Eqs. (10), (13), and (16), and found no step that reduces by construction to its own inputs or to a self-citation. The central theoretical justification, Theorem 4.1, is explicitly built on external results in constrained Langevin sampling, references [12, 13, 14, 15], none of which are authored by the present paper's authors; the paper is therefore not importing a uniqueness or convergence theorem from itself. The experimental claims are evaluated against standard benchmarks and baselines, with no fitted parameter being relabeled as a prediction; constraint violations and computation times are measured on trajectories generated by the proposed algorithms, not derived from the same numbers used to fit the score. The one substantive technical concern, that Eq. (4) is not algebraically identical to Eq. (6) under the stated Tweedie substitution, is a correctness or approximation issue rather than a circularity: Eq. (6) is presented as a reformulation of Eq. (4), not assumed as an input, so the derivation chain may be flawed but it is not circular. I therefore find no self-referential reduction and assign a circularity score of 0.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The framework relies on standard diffusion score estimation plus unstated regularity conditions from prior constrained sampling theory. No new physical entities are introduced. The main load-bearing assumptions are the Langevin equivalence for the DDPM update, the existence of an inverse dynamics model, and the differentiability of constraints.

free parameters (3)
  • alpha (DCBF decay coefficient) = not reported
    Chosen by hand in Eq. (19); controls how aggressively the state remains inside the safe set and affects violation rates.
  • eta_lambda (dual step size) = not reported
    Step size for the Lagrange multiplier update in Eq. (14); must be tuned for stable convergence.
  • rho_0 and c (ALM penalty schedule) = not reported
    Initial penalty and growth factor in Section 4.3; these affect how strictly inequality constraints are enforced.
assumptions (5)
  • domain assumption The discrete DDPM reverse update (Eq. 4) is equivalent to a Langevin sampling step with step size beta_t/2 (Eq. 6).
    The 1/sqrt(1-beta_t) scaling and the exact score are dropped; the paper claims consistency only in the T to infinity limit.
  • domain assumption An inverse dynamics model exists and is trainable, and there are no input constraints.
    Stated in Section 5; action generation during online implementation relies on this existence.
  • domain assumption Constraint functions g are differentiable or replaced by the softplus surrogate.
    Remark 1 introduces a differentiable approximation for non-smooth constraints, which is needed for gradient-based updates.
  • ad hoc to paper Mild conditions for Theorem 4.1 hold, such as boundedness, Lipschitz continuity, and proper step sizes.
    The conditions are not enumerated in the main text; the convergence theorem rests on this unstated regularity.
  • domain assumption The score function of the pre-trained diffusion model approximates the true score of the data distribution well enough for constraint propagation.
    All constrained updates use the neural score; score error propagates into the quality of constraint satisfaction.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Constrained Diffusers for Safe Planning and Control." pith.science (2026). https://pith.science/paper/JFAOXD6O

@misc{pith2026250612544,
  author       = {Pith},
  title        = {Pith review of: Constrained Diffusers for Safe Planning and Control},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JFAOXD6O}},
  note         = {Machine review of arXiv:2506.12544}
}
read the original abstract

Diffusion models have shown remarkable potential in planning and control tasks due to their ability to represent multimodal distributions over actions and trajectories. However, ensuring safety under constraints remains a critical challenge for diffusion models. This paper proposes Constrained Diffusers, a novel framework that incorporates constraints into pre-trained diffusion models without retraining or architectural modifications. Inspired by constrained optimization, we apply a constrained Langevin sampling mechanism for the reverse diffusion process that jointly optimizes the trajectory and realizes constraint satisfaction through three iterative algorithms: projected method, primal-dual method and augmented Lagrangian approaches. In addition, we incorporate discrete control barrier functions as constraints for constrained diffusers to guarantee safety in online implementation. Experiments in Maze2D, locomotion, and pybullet ball running tasks demonstrate that our proposed methods achieve constraint satisfaction with less computation time, and are competitive to existing methods in environments with static and time-varying constraints.

Figures

Figures reproduced from arXiv: 2506.12544 by the authors.

Figure 1
Figure 1. Constrained sampling trajectories through (a) projected, (b) primal-dual and (c) augmented [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overall Framework. We use expert trajectories to train the diffuser, apply constrained [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Trajectories generated by Constrained Dif [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: PybulletBallRunning tasks with the black-white ball and a red moving obstacles ball. Finally, we evaluate the adaptability of our proposed algorithm in the PyBullet SafetyBallRun environment, where safety constraints varies over time. In this setup, we have a obstacle …

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. TPD: Temporal Prior Decoupling for Text-to-Video Diffusion Models

    cs.CV 2026-07 conditional novelty 6.5 of 10

    TPD restores suppressed late-segment events in text-to-video diffusion by projecting classifier-free guidance onto a frame- and timestep-selective lower bound along a temporal-counterfactual direction.

  2. Motion Planning with Model-Based Diffusion via Constraint Optimization and Adaptive Scheduling

    cs.RO 2026-07 conditional novelty 6.0 of 10

    MD-COAS unifies inexact augmented-Lagrangian soft constraints with convex-feasible-set hard projection and adaptively schedules them during model-based diffusion, improving safe and successful planning in non-convex e...

Reference graph

Works this paper leans on

43 extracted references · 19 canonical work pages · cited by 2 Pith papers

  1. [1]

    Deep unsupervised learning using nonequilibrium thermodynamics

    Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Surya Ganguli. Deep unsupervised learning using nonequilibrium thermodynamics. In International conference on machine learning, pages 2256–2265. pmlr, 2015

    2015

  2. [2]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020

    2020

  3. [3]

    Planning with diffusion for flexible behavior synthesis

    Michael Janner, Yilun Du, Joshua B Tenenbaum, and Sergey Levine. Planning with diffusion for flexible behavior synthesis. arXiv preprint arXiv:2205.09991, 2022

    arXiv 2022

  4. [4]

    Is conditional generative modeling all you need for decision-making? arXiv preprint arXiv:2211.15657, 2022

    Anurag Ajay, Yilun Du, Abhi Gupta, Joshua Tenenbaum, Tommi Jaakkola, and Pulkit Agrawal. Is conditional generative modeling all you need for decision-making? arXiv preprint arXiv:2211.15657, 2022

    arXiv 2022

  5. [5]

    Diffusion models beat gans on image synthesis

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis. Advances in neural information processing systems, 34:8780–8794, 2021

    2021

  6. [6]

    Classifier-free diffusion guidance

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance. arXiv preprint arXiv:2207.12598, 2022

    arXiv 2022

  7. [7]

    Diffusion policy policy optimization

    Allen Z Ren, Justin Lidard, Lars L Ankile, Anthony Simeonov, Pulkit Agrawal, Anirudha Majumdar, Benjamin Burchfiel, Hongkai Dai, and Max Simchowitz. Diffusion policy policy optimization. arXiv preprint arXiv:2409.00588, 2024

    arXiv 2024

  8. [8]

    Safe offline reinforcement learning with feasibility-guided diffusion model

    Yinan Zheng, Jianxiong Li, Dongjie Yu, Yujie Yang, Shengbo Eben Li, Xianyuan Zhan, and Jingjing Liu. Safe offline reinforcement learning with feasibility-guided diffusion model. arXiv preprint arXiv:2401.10700, 2024

    arXiv 2024

Show all 43 references
  1. [9]

    Generative modeling by estimating gradients of the data distribution

    Yang Song and Stefano Ermon. Generative modeling by estimating gradients of the data distribution. Advances in neural information processing systems, 32, 2019

    2019

  2. [10]

    Sampling as optimization in the space of measures: The langevin dynamics as a composite optimization problem

    Andre Wibisono. Sampling as optimization in the space of measures: The langevin dynamics as a composite optimization problem. In Conference on learning theory, pages 2093–3027. PMLR, 2018

    2018

  3. [11]

    Stochastic Approximation on Riemannian Manifolds and the Space of Measures

    Mohammad R Karimi Jaghargh. Stochastic Approximation on Riemannian Manifolds and the Space of Measures. PhD thesis, ETH Zurich, 2024

    2024

  4. [12]

    Sampling from a log-concave distribution with projected langevin monte carlo

    Sébastien Bubeck, Ronen Eldan, and Joseph Lehec. Sampling from a log-concave distribution with projected langevin monte carlo. Discrete & Computational Geometry, 59:757–783, 2018

    2018

  5. [13]

    Constrained sampling with primal-dual langevin monte carlo

    Luiz Chamon, Mohammad Reza Karimi Jaghargh, and Anna Korba. Constrained sampling with primal-dual langevin monte carlo. Advances in Neural Information Processing Systems, 37:29285–29323, 2024

    2024

  6. [14]

    Sampling with trusthworthy constraints: A variational gradient framework

    Xingchao Liu, Xin Tong, and Qiang Liu. Sampling with trusthworthy constraints: A variational gradient framework. Advances in neural information processing systems, 34:23557–23568, 2021. 10

    2021

  7. [15]

    Springer, 2008

    Vivek S Borkar and Vivek S Borkar.Stochastic approximation: a dynamical systems viewpoint, volume 9. Springer, 2008

    2008

  8. [16]

    Score-based generative modeling through stochastic differential equations

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations. arXiv preprint arXiv:2011.13456, 2020

    2011 arXiv

  9. [17]

    Understanding diffusion models: A unified perspective

    Calvin Luo. Understanding diffusion models: A unified perspective. arXiv preprint arXiv:2208.11970, 2022

    2022 arXiv

  10. [18]

    Control barrier functions: Theory and applications

    Aaron D Ames, Samuel Coogan, Magnus Egerstedt, Gennaro Notomista, Koushil Sreenath, and Paulo Tabuada. Control barrier functions: Theory and applications. In 2019 18th European control conference (ECC), pages 3420–3431. Ieee, 2019

    2019

  11. [19]

    Discrete control barrier functions for safety-critical control of discrete systems with application to bipedal robot navigation

    Ayush Agrawal and Koushil Sreenath. Discrete control barrier functions for safety-critical control of discrete systems with application to bipedal robot navigation. InRobotics: Science and Systems, volume 13, pages 1–10. Cambridge, MA, USA, 2017

  12. [20]

    Nonlinear Programming

    Dimitri Bertsekas. Nonlinear Programming. Athena Scientific, 2003

    2003

  13. [21]

    Bonnans and Alexander Shapiro

    J. Bonnans and Alexander Shapiro. Perturbation Analysis of Optimization Problems. Springer, 2000

    2000

  14. [22]

    Constrained optimization and Lagrange multiplier methods

    Dimitri P Bertsekas. Constrained optimization and Lagrange multiplier methods. Academic press, 2014

    2014

  15. [23]

    Augmented lagrangian method for instantaneously constrained reinforcement learning problems

    Jingqi Li, David Fridovich-Keil, Somayeh Sojoudi, and Claire J Tomlin. Augmented lagrangian method for instantaneously constrained reinforcement learning problems. In 2021 60th IEEE Conference on Decision and Control (CDC), pages 2982–2989. IEEE, 2021

    2021

  16. [24]

    Safediffuser: Safe planning with diffusion probabilistic models

    Wei Xiao, Tsun-Hsuan Wang, Chuang Gan, and Daniela Rus. Safediffuser: Safe planning with diffusion probabilistic models. arXiv preprint arXiv:2306.00148, 2023

    2023 arXiv

  17. [25]

    Gymnasium: A standard interface for reinforcement learning environments

    Mark Towers, Ariel Kwiatkowski, Jordan Terry, John U Balis, Gianluca De Cola, Tristan Deleu, Manuel Goulão, Andreas Kallinteris, Markus Krimmel, Arjun KG, et al. Gymnasium: A standard interface for reinforcement learning environments. arXiv preprint arXiv:2407.17032, 2024

    2024 arXiv

  18. [26]

    Glide: Towards photorealistic image generation and editing with text-guided diffusion models

    Alexander Quinn Nichol, Prafulla Dhariwal, Aditya Ramesh, Pranav Shyam, Pamela Mishkin, Bob Mcgrew, Ilya Sutskever, and Mark Chen. Glide: Towards photorealistic image generation and editing with text-guided diffusion models. In International Conference on Machine Learning, pag...

    2022

  19. [27]

    Photorealistic text-to- image diffusion models with deep language understanding

    Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily L Denton, Kamyar Ghasemipour, Raphael Gontijo Lopes, Burcu Karagol Ayan, Tim Salimans, et al. Photorealistic text-to- image diffusion models with deep language understanding. Advances in neural informatio...

    2022

  20. [28]

    Reverse-time diffusion equation models

    Brian DO Anderson. Reverse-time diffusion equation models. Stochastic Processes and their Applications, 12(3):313–326, 1982

    1982

  21. [29]

    Diffusion policy: Visuomotor policy learning via action diffusion

    Cheng Chi, Zhenjia Xu, Siyuan Feng, Eric Cousineau, Yilun Du, Benjamin Burchfiel, Russ Tedrake, and Shuran Song. Diffusion policy: Visuomotor policy learning via action diffusion. The International Journal of Robotics Research, page 02783649241273668, 2023

    2023

  22. [30]

    Model-based diffusion for trajectory optimization

    Chaoyi Pan, Zeji Yi, Guanya Shi, and Guannan Qu. Model-based diffusion for trajectory optimization. Advances in Neural Information Processing Systems, 37:57914–57943, 2024

    2024

  23. [31]

    Diffusolve: Diffusion-based solver for non-convex trajectory optimization

    Anjian Li, Zihan Ding, Adji Bousso Dieng, and Ryne Beeson. Diffusolve: Diffusion-based solver for non-convex trajectory optimization. arXiv preprint arXiv:2403.05571, 2024

    2024 arXiv

  24. [32]

    Diffusion models for constrained domains

    Nic Fishman, Leo Klarner, Valentin De Bortoli, Emile Mathieu, and Michael Hutchinson. Diffusion models for constrained domains. arXiv preprint arXiv:2304.05364, 2023

    2023 arXiv

  25. [33]

    Efficient constrained sampling via the mirror-langevin algorithm

    Kwangjun Ahn and Sinho Chewi. Efficient constrained sampling via the mirror-langevin algorithm. Advances in Neural Information Processing Systems, 34:28405–28418, 2021

    2021

  26. [34]

    Sampling with riemannian hamiltonian monte carlo in a constrained space

    Yunbum Kook, Yin-Tat Lee, Ruoqi Shen, and Santosh Vempala. Sampling with riemannian hamiltonian monte carlo in a constrained space. Advances in Neural Information Processing Systems, 35:31684–31696, 2022

    2022

  27. [35]

    Penalized langevin and hamiltonian monte carlo algorithms for constrained sampling

    Mert Gürbüzbalaban, Yuanhan Hu, and Lingjiong Zhu. Penalized langevin and hamiltonian monte carlo algorithms for constrained sampling. arXiv preprint arXiv:2212.00570, 2022. 11

    2022 arXiv

  28. [36]

    Safe learning in robotics: From learning-based control to safe reinforcement learning

    Lukas Brunke, Melissa Greeff, Adam W Hall, Zhaocong Yuan, Siqi Zhou, Jacopo Panerati, and Angela P Schoellig. Safe learning in robotics: From learning-based control to safe reinforcement learning. Annual Review of Control, Robotics, and Autonomous Systems, 5(1):411–444, 2022

    2022

  29. [37]

    Comparing forward and backward reachability as tools for safety analysis

    Ian M Mitchell. Comparing forward and backward reachability as tools for safety analysis. InInternational Workshop on Hybrid Systems: Computation and Control, pages 428–443. Springer, 2007

    2007

  30. [38]

    Model predictive control: Theory and practice—a survey

    Carlos E Garcia, David M Prett, and Manfred Morari. Model predictive control: Theory and practice—a survey. Automatica, 25(3):335–348, 1989

    1989

  31. [39]

    A comprehensive survey on safe reinforcement learning

    Javier Garcıa and Fernando Fernández. A comprehensive survey on safe reinforcement learning. Journal of Machine Learning Research, 16(1):1437–1480, 2015

    2015

  32. [40]

    Constrained policy optimization

    Joshua Achiam, David Held, Aviv Tamar, and Pieter Abbeel. Constrained policy optimization. In International conference on machine learning, pages 22–31. PMLR, 2017

    2017

  33. [41]

    Benchmarking safe exploration in deep reinforcement learning

    Alex Ray, Joshua Achiam, and Dario Amodei. Benchmarking safe exploration in deep reinforcement learning

  34. [42]

    Optnet: Differentiable optimization as a layer in neural networks

    Brandon Amos and J Zico Kolter. Optnet: Differentiable optimization as a layer in neural networks. In International conference on machine learning, pages 136–145. PMLR, 2017

    2017

  35. [43]

    Risk-constrained rein- forcement learning with percentile risk criteria

    Yinlam Chow, Mohammad Ghavamzadeh, Lucas Janson, and Marco Pavone. Risk-constrained rein- forcement learning with percentile risk criteria. Journal of Machine Learning Research, 18(167):1–51, 2018. 12

    2018

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.