Ambient Diffusion Policy: Imitation Learning from Suboptimal Data in Robotics

• URL: http://arxiv.org/abs/2606.12365v1
• Date: Wed, 10 Jun 2026 17:34:12 GMT
• Title: Ambient Diffusion Policy: Imitation Learning from Suboptimal Data in Robotics
• Abstract summary: Ambient Diffusion Policy is a principled method for imitation learning from suboptimal data in robotics.<n>Our method extracts only the useful features by introducing a new axis to co-training in robotics: noise-dependent data usage.<n>Our experiments validate Ambient Diffusion Policy on four types of suboptimal action data across six tasks.
• Score: 29.36264775500445
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose Ambient Diffusion Policy, a simple and principled method for imitation learning from suboptimal data in robotics. High-quality, task-specific robot data is expensive and time-consuming to collect, while suboptimal datasets with lower-quality or out-of-distribution demonstrations are abundant. Existing methods that co-train on both data sources in robotics often fail to separate the meaningful and the harmful features in the suboptimal samples. In contrast, our method extracts only the useful features by introducing a new axis to co-training in robotics: noise-dependent data usage. Ambient Diffusion Policy restricts the contribution of suboptimal data during training to only the high and low diffusion times. To rigorously justify our approach, we first observe that robot action data exhibits a spectral power law. This induces two important properties on the optimal Diffusion Policy that we exploit: a global-to-local hierarchy and locality. We theoretically formalize this discussion using a simplified model. Our experiments validate Ambient Diffusion Policy on four types of suboptimal action data (noisy trajectories, sim-to-real gap, task mismatch, and large-scale data mixtures) across six tasks. The results show that it effectively learns from arbitrary sources of suboptimal data. Notably, it outperforms existing co-training baselines by up to 33% when scaled to Open X-Embodiment - a large dataset with heterogeneous data quality and unstructured distribution shifts. Overall, Ambient Diffusion Policy increases the utility of suboptimal demonstrations and expands the set of usable data sources in robotics.

Related papers

• Studying the Role of Synthetic Data for Machine Learning-based Wireless Networks Traffic Forecasting [1.1699027359021665]
  This paper proposes a novel method to generate synthetic data, based on first-order auto-regressive noise statistics, for large-scale Wi-Fi deployments.<n> Experimental results show that ML models trained on synthetic data achieve Mean Absolute Error (MAE) values within 10 to 15 of those obtained using real data.<n>When generalization is required, synthetic-data-trained models improve prediction accuracy by up to 50 percent compared to real-data-trained baselines.
  arXiv  Detail & Related papers  (2026-01-12T15:27:55Z)
• Synthetic Industrial Object Detection: GenAI vs. Feature-Based Methods [5.278929538141005]
  We benchmark a range of domain randomization (DR) and domain adaptation (DA) techniques, including feature-based methods, generative AI (GenAI) and classical rendering approaches.<n>Our evaluation focuses on the effectiveness and efficiency of low-level and high-level feature alignment, as well as a controlled diffusion-based DA method guided by prompts generated from real-world contexts.<n>Results show that if render-based data with enough variability is available as seed, simpler feature-based methods, such as brightness-based and perceptual hashing filtering, outperform more complex GenAI-based approaches in both accuracy and resource efficiency
  arXiv  Detail & Related papers  (2025-11-28T14:51:08Z)
• Trajectory prediction for heterogeneous agents: A performance analysis on small and imbalanced datasets [11.00008290019651]
  Class-conditioned trajectory prediction is an appealing way to reduce forecast uncertainty and get more accurate predictions.<n>This is hardly explored in the prior art, especially for mobile robots and in limited data applications.<n>We propose a set of conditional pattern-based and efficient deep learning-based baselines, and evaluate their performance on robotics and outdoors datasets.
  arXiv  Detail & Related papers  (2025-10-04T11:02:21Z)
• Filtering with Confidence: When Data Augmentation Meets Conformal Prediction [10.431187957711206]
  We propose a principled data filtering framework for synthetic data augmentation.<n>We show consistent performance improvements of up to 40% in F1 score over unaugmented baselines.
  arXiv  Detail & Related papers  (2025-09-25T19:42:02Z)
• Efficient Federated Learning with Heterogeneous Data and Adaptive Dropout [62.73150122809138]
  Federated Learning (FL) is a promising distributed machine learning approach that enables collaborative training of a global model using multiple edge devices.<n>We propose the FedDHAD FL framework, which comes with two novel methods: Dynamic Heterogeneous model aggregation (FedDH) and Adaptive Dropout (FedAD)<n>The combination of these two methods makes FedDHAD significantly outperform state-of-the-art solutions in terms of accuracy (up to 6.7% higher), efficiency (up to 2.02 times faster), and cost (up to 15.0% smaller)
  arXiv  Detail & Related papers  (2025-07-14T16:19:00Z)
• DataMIL: Selecting Data for Robot Imitation Learning with Datamodels [77.48472034791213]
  We introduce DataMIL, a policy-driven data selection framework built on the datamodels paradigm.<n>Unlike standard practices that filter data using human notions of quality, DataMIL directly optimize data selection for task success.<n>We validate our approach on a suite of more than 60 simulation and real-world manipulation tasks.
  arXiv  Detail & Related papers  (2025-05-14T17:55:10Z)
• Efficient Data Collection for Robotic Manipulation via Compositional Generalization [70.76782930312746]
  We show that policies can compose environmental factors from their data to succeed when encountering unseen factor combinations. We propose better in-domain data collection strategies that exploit composition. We provide videos at http://iliad.stanford.edu/robot-data-comp/.
  arXiv  Detail & Related papers  (2024-03-08T07:15:38Z)
• STAR: Boosting Low-Resource Information Extraction by Structure-to-Text Data Generation with Large Language Models [56.27786433792638]
  STAR is a data generation method that leverages Large Language Models (LLMs) to synthesize data instances. We design fine-grained step-by-step instructions to obtain the initial data instances. Our experiments show that the data generated by STAR significantly improve the performance of low-resource event extraction and relation extraction tasks.
  arXiv  Detail & Related papers  (2023-05-24T12:15:19Z)
• Synthetic data, real errors: how (not) to publish and use synthetic data [86.65594304109567]
  We show how the generative process affects the downstream ML task. We introduce Deep Generative Ensemble (DGE) to approximate the posterior distribution over the generative process model parameters.
  arXiv  Detail & Related papers  (2023-05-16T07:30:29Z)
• DRFLM: Distributionally Robust Federated Learning with Inter-client Noise via Local Mixup [58.894901088797376]
  federated learning has emerged as a promising approach for training a global model using data from multiple organizations without leaking their raw data. We propose a general framework to solve the above two challenges simultaneously. We provide comprehensive theoretical analysis including robustness analysis, convergence analysis, and generalization ability.
  arXiv  Detail & Related papers  (2022-04-16T08:08:29Z)

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.