面向流形感知扩散规划的局部流形近似与投影
Local Manifold Approximation and Projection for Manifold-Aware Diffusion Planning
AI总结：
针对扩散规划中引导不准确导致不可行轨迹的问题，本文提出免训练的局部流形近似与投影方法LoMAP，将采样投影到离线数据低秩子空间，并在标准离线强化学习基准上验证其有效性。
摘要：
近年来，基于扩散的生成建模在利用离线数据集处理长时程、稀疏奖励任务方面展现出显著潜力。尽管这些方法取得了有希望的结果，但由于产生不可行轨迹的固有随机风险，其可靠性仍不一致，限制了它们在安全关键型应用中的适用性。我们将这些失败的主要原因归结为采样过程中的不准确引导，并通过推导引导差距的下界证明了流形偏差的存在。为应对这一挑战，我们提出了局部流形近似与投影（LoMAP），这是一种免训练方法，将引导后的样本投影到从离线数据集近似的低秩子空间上，从而防止不可行轨迹的生成。我们在涉及具有挑战性的长时程规划的标准离线强化学习基准上验证了我们的方法。此外，我们表明，作为一个独立模块，LoMAP 可以集成到分层扩散规划器中，从而提供进一步的性能提升。
英文摘要：
Recent advances in diffusion-based generative modeling have demonstrated significant promise in tackling long-horizon, sparse-reward tasks by leveraging offline datasets. While these approaches have achieved promising results, their reliability remains inconsistent due to the inherent stochastic risk of producing infeasible trajectories, limiting their applicability in safety-critical applications. We identify that the primary cause of these failures is inaccurate guidance during the sampling procedure, and demonstrate the existence of manifold deviation by deriving a lower bound on the guidance gap. To address this challenge, we propose Local Manifold Approximation and Projection (LoMAP), a training-free method that projects the guided sample onto a low-rank subspace approximated from offline datasets, preventing infeasible trajectory generation. We validate our approach on standard offline reinforcement learning benchmarks that involve challenging long-horizon planning. Furthermore, we show that, as a standalone module, LoMAP can be incorporated into the hierarchical diffusion planner, providing further performance enhancements.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
