mlsanityData to 7 Oct 2026

Bayesian Disturbance Injection: Robust Imitation Learning of Flexible Policies for Robot Manipulation

arXiv 2211.03393 ↗First posted 7 Nov 2022cs.ROcited by 8

On the map: Offline RL and imitation learning from data → Imitation learning methods

Abstract (arXiv)

Humans demonstrate a variety of interesting behavioral characteristics when performing tasks, such as selecting between seemingly equivalent optimal actions, performing recovery actions when deviating from the optimal trajectory, or moderating actions in response to sensed risks. However, imitation learning, which attempts to teach robots to perform these same tasks from observations of human demonstrations, often fails to capture such behavior. Specifically, commonly used learning algorithms embody inherent contradictions between the learning assumptions (e.g., single optimal action) and actual human behavior (e.g., multiple optimal actions), thereby limiting robot generalizability, applicability, and demonstration feasibility. To address this, this paper proposes designing imitation learning algorithms with a focus on utilizing human behavioral characteristics, thereby embodying principles for capturing and exploiting actual demonstrator behavioral characteristics. This paper presents the first imitation learning framework, Bayesian Disturbance Injection (BDI), that typifies human behavioral characteristics by incorporating model flexibility, robustification, and risk sensitivity. Bayesian inference is used to learn flexible non-parametric multi-action policies, while simultaneously robustifying policies by injecting risk-sensitive disturbances to induce human recovery action and ensuring demonstration feasibility. Our method is evaluated through risk-sensitive simulations and real-robot experiments (e.g., table-sweep task, shaft-reach task and shaft-insertion task) using the UR5e 6-DOF robotic arm, to demonstrate the improved characterisation of behavior. Results show significant improvement in task performance, through improved flexibility, robustness as well as demonstration feasibility.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 15 of its 57 references that are in the map. First posted on arXiv 7 Nov 2022.

Builds on

15 of 57 references.

Closest ancestors · 3

Same direction, newest first

  1. 2022Disturbance-Injected Robust Imitation Learning with Task Achievement9 May
  2. 2021Bayesian Disturbance Injection: Robust Imitation Learning of Flexible Policies25 Mar
  3. 2018An Algorithmic Perspective on Imitation Learning16 Nov

Other references · 12

  1. 2016End to End Learning for Self-Driving Carscited by 1,000
  2. 2016Learning Hand-Eye Coordination for Robotic Grasping with Deep Learning and Large-Scale Data Collectioncited by 312
  3. 2017Deep Imitation Learning for Complex Manipulation Tasks from Virtual Reality Teleoperationcited by 286
  4. 2017DART: Noise Injection for Robust Imitation Learningcited by 187
  5. 2017Vision-Based Multi-Task Manipulation for Inexpensive Robots Using End-To-End Learning from Demonstrationcited by 100
  6. 2020Learning from Suboptimal Demonstration via Self-Supervised Reward Regressioncited by 84
  7. 2014Fastfood: Approximate Kernel Expansions in Loglinear Timecited by 72
  8. 2017A survey of robot learning from demonstrations for Human-Robot Collaborationcited by 38
  9. 2017Kernelized Movement Primitivescited by 37
  10. 2019Regression Planning Networkscited by 32
  11. 2018MPC-Inspired Neural Network Policies for Sequential Decision Makingcited by 13
  12. 2019Learning a Multi-Modal Policy via Imitating Demonstrations with Mixed Behaviorscited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 6; 2026: 1.

8 papers.

  1. 2024Leveraging Demonstrator-perceived Precision for Safe Interactive Imitation Learning of Clearance-limited Taskscited by 5
  2. 2025Robust Instant Policy: Leveraging Student's t-Regression Model for Robust In-context Imitation Learning of Robot Manipulationcited by 2
  3. 2025Self-Augmented Robot Trajectory: Efficient Imitation Learning via Safe Self-augmentation with Demonstrator-annotated Precisioncited by 2
  4. 2025Feasibility-aware Imitation Learning from Observations through a Hand-mounted Demonstration Interfacecited by 1
  5. 2025Exploiting Policy Idling for Dexterous Manipulationcited by 1
  6. 2025Composite Gaussian Processes Flows for Learning Discontinuous Multimodal Policiescited by 0
  7. 2025CubeDAgger: Interactive Imitation Learning for Dynamic Systems with Efficient yet Low-risk Interactioncited by 0
  8. 2026Feasibility-aware Imitation Learning from Observation with Multimodal Feedbackcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Bayesian Optimization for Sample-Efficient Policy Improvement in Robotic Manipulation ↗
  2. 2026It's Not Just More Demos: Counterfactual Action Sensitivity Coverage for Data-Efficient Robust Robot Imitation
  3. 2024A context-adaptive policy framework for robust and reactive robotic manipulation via uncertainty-aware imitation learning
  4. 2025Robotic Manipulation via Imitation Learning: Taxonomy, Evolution, Benchmark, and Challenges
  5. 2025RISE: Robust Imitation through Stochastic Encoding
  6. 2019A Linearly Constrained Nonparametric Framework for Imitation Learning
  7. 2023Aligning Human Intent from Imperfect Demonstrations with Confidence-based Inverse soft-Q Learning
  8. 2024VITAL: Interactive Few-Shot Imitation Learning via Visual Human-in-the-Loop Corrections