Accelerating trajectory optimization with Sobolev-trained diffusion policies

• URL: http://arxiv.org/abs/2604.19011v1
• Date: Tue, 21 Apr 2026 03:00:52 GMT
• Title: Accelerating trajectory optimization with Sobolev-trained diffusion policies
• Abstract summary: Trajectory Optimization (TO) solvers exploit known system dynamics to compute locally optimal trajectories.<n>A downside is that each new problem instance is solved independently.<n>To improve efficiency, a natural approach is to warm-start TO with initial guesses produced by a learned policy.
• Score: 12.684196947818103
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Trajectory Optimization (TO) solvers exploit known system dynamics to compute locally optimal trajectories through iterative improvements. A downside is that each new problem instance is solved independently; therefore, convergence speed and quality of the solution found depend on the initial trajectory proposed. To improve efficiency, a natural approach is to warm-start TO with initial guesses produced by a learned policy trained on trajectories previously generated by the solver. Diffusion-based policies have recently emerged as expressive imitation learning models, making them promising candidates for this role. Yet, a counterintuitive challenge comes from the local optimality of TO demonstrations: when a policy is rolled out, small non-optimal deviations may push it into situations not represented in the training data, triggering compounding errors over long horizons. In this work, we focus on learning-based warm-starting for gradient-based TO solvers that also provide feedback gains. Exploiting this specificity, we derive a first-order loss for Sobolev learning of diffusion-based policies using both trajectories and feedback gains. Through comprehensive experiments, we demonstrate that the resulting policy avoids compounding errors, and so can learn from very few trajectories to provide initial guesses reducing solving time by $2\times$ to $20 \times$. Incorporating first-order information enables predictions with fewer diffusion steps, reducing inference latency.

Related papers

• Zero-Shot Off-Policy Learning [9.729890516322781]
  Off-policy learning methods seek to derive an optimal policy directly from a fixed dataset of prior interactions.<n>In this work, we address the off-policy problem in a zero-shot setting by discovering a theoretical connection of successor measures to stationary density ratios.<n>Our algorithm can infer optimal importance sampling ratios, effectively performing a stationary distribution correction with an optimal policy for any task on the fly.
  arXiv  Detail & Related papers  (2026-02-02T11:06:31Z)
• Q-learning with Adjoint Matching [58.78551025170267]
  We propose Q-learning with Adjoint Matching (QAM), a novel TD-based reinforcement learning (RL) algorithm.<n>QAM sidesteps two challenges by leveraging adjoint matching, a recently proposed technique in generative modeling.<n>It consistently outperforms prior approaches on hard, sparse reward tasks in both offline and offline-to-online RL.
  arXiv  Detail & Related papers  (2026-01-20T18:45:34Z)
• A Dual Perspective on Decision-Focused Learning: Scalable Training via Dual-Guided Surrogates [1.7100385719232911]
  Decision-focused learning trains models with awareness of how predictions refreshes, improving the performance of downstream decisions.<n>Despite its promise, scaling is challenging: state-of-the-art methods either differentiate through a solver or rely on task-specific surrogates.<n>In this paper, we leverage dual variables to shape learning and introduce Dual-Guided Loss (DGL)<n>DGL matches or exceeds state-of-the-art DFL methods while using far fewer calls and substantially less training time.
  arXiv  Detail & Related papers  (2025-11-07T01:15:15Z)
• ResAD: Normalized Residual Trajectory Modeling for End-to-End Autonomous Driving [64.42138266293202]
  ResAD is a Normalized Residual Trajectory Modeling framework.<n>It reframes the learning task to predict the residual deviation from an inertial reference.<n>On the NAVSIM benchmark, ResAD achieves a state-of-the-art PDMS of 88.6 using a vanilla diffusion policy.
  arXiv  Detail & Related papers  (2025-10-09T17:59:36Z)
• Online Decision-Focused Learning [74.3205104323777]
  Decision-focused learning (DFL) is an increasingly popular paradigm for training models whose predictive outputs are used in decision-making tasks.<n>In this paper, we regularize the objective function to make it different and investigate how to overcome nonoptimality function.<n>We also showcase the effectiveness of our algorithms on a knapsack experiment, where they outperform two standard benchmarks.
  arXiv  Detail & Related papers  (2025-05-19T10:40:30Z)
• DeepONet as a Multi-Operator Extrapolation Model: Distributed Pretraining with Physics-Informed Fine-Tuning [6.635683993472882]
  We propose a novel fine-tuning method to achieve multi-operator learning. Our approach combines distributed learning to integrate data from various operators in pre-training, while physics-informed methods enable zero-shot fine-tuning.
  arXiv  Detail & Related papers  (2024-11-11T18:58:46Z)
• CGD: Constraint-Guided Diffusion Policies for UAV Trajectory Planning [26.10588918124538]
  A successful strategy to reduce computation time involves using Imitation Learning (IL) to develop fast neural network (NN) policies from experts. Although the resulting NN policies are effective at quickly generating trajectories similar to those from the expert, their output does not explicitly account for dynamic feasibility. We propose Constraint-Guided Diffusion (CGD), a novel IL-based approach to trajectory planning.
  arXiv  Detail & Related papers  (2024-05-02T21:50:26Z)
• Learning Constrained Optimization with Deep Augmented Lagrangian Methods [54.22290715244502]
  A machine learning (ML) model is trained to emulate a constrained optimization solver. This paper proposes an alternative approach, in which the ML model is trained to predict dual solution estimates directly. It enables an end-to-end training scheme is which the dual objective is as a loss function, and solution estimates toward primal feasibility, emulating a Dual Ascent method.
  arXiv  Detail & Related papers  (2024-03-06T04:43:22Z)
• Imitating, Fast and Slow: Robust learning from demonstrations via decision-time planning [96.72185761508668]
  Planning at Test-time (IMPLANT) is a new meta-algorithm for imitation learning. We demonstrate that IMPLANT significantly outperforms benchmark imitation learning approaches on standard control environments.
  arXiv  Detail & Related papers  (2022-04-07T17:16:52Z)
• Deep learning: a statistical viewpoint [120.94133818355645]
  Deep learning has revealed some major surprises from a theoretical perspective. In particular, simple gradient methods easily find near-perfect solutions to non-optimal training problems. We conjecture that specific principles underlie these phenomena.
  arXiv  Detail & Related papers  (2021-03-16T16:26:36Z)
• Off-Policy Imitation Learning from Observations [78.30794935265425]
  Learning from Observations (LfO) is a practical reinforcement learning scenario from which many applications can benefit. We propose a sample-efficient LfO approach that enables off-policy optimization in a principled manner. Our approach is comparable with state-of-the-art locomotion in terms of both sample-efficiency and performance.
  arXiv  Detail & Related papers  (2021-02-25T21:33:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.