mlsanityData to 14 Oct 2026

Hierarchical Policy Blending As Optimal Transport

arXiv 2212.01938 ↗First posted 4 Dec 2022cs.RO cs.LG cs.SYcited by 4

Not on the map. Why?

Abstract (arXiv)

We present hierarchical policy blending as optimal transport (HiPBOT). HiPBOT hierarchically adjusts the weights of low-level reactive expert policies of different agents by adding a look-ahead planning layer on the parameter space. The high-level planner renders policy blending as unbalanced optimal transport consolidating the scaling of the underlying Riemannian motion policies. As a result, HiPBOT effectively decides the priorities between expert policies and agents, ensuring the task's success and guaranteeing safety. Experimental results in several application scenarios, from low-dimensional navigation to high-dimensional whole-body control, show the efficacy and efficiency of HiPBOT. Our method outperforms state-of-the-art baselines -- either adopting probabilistic inference or defining a tree structure of experts -- paving the way for new applications of optimal transport to robot control. More material at https://sites.google.com/view/hipobot

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 31 of its 69 references that are in the map. First posted on arXiv 4 Dec 2022.

Builds on

31 of 69 references.

References · 31

  1. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  2. 2018Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Reviewcited by 482
  3. 2017Near-linear time approximation algorithms for optimal transport via Sinkhorn iterationcited by 223
  4. 2015Learning with a Wasserstein Losscited by 196
  5. 2022Watch and Match: Supercharging Imitation with Regularized Optimal Transportcited by 67
  6. 2020Long-Horizon Visual Planning with Goal-Conditioned Hierarchical Predictorscited by 62
  7. 2020On Unbalanced Optimal Transport: An Analysis of Sinkhorn Algorithmcited by 61
  8. 2020Neural Dynamic Policies for End-to-End Sensorimotor Learningcited by 58
  9. 2017Continuous-Time Gaussian Process Motion Planning via Probabilistic Inferencecited by 49
  10. 2020ReLMoGen: Leveraging Motion Generation in Reinforcement Learning for Mobile Manipulationcited by 49
  11. 2021STORM: An Integrated Framework for Fast Joint-Space Model-Predictive Control for Reactive Manipulationcited by 47
  12. 2018Riemannian Motion Policiescited by 46
19 more
  1. 2021Autonomous Reinforcement Learning via Subgoal Curriculacited by 33
  2. 2021Specializing Versatile Skill Libraries using Local Mixture of Expertscited by 24
  3. 2022Robot Learning of Mobile Manipulation with Reachability Behavior Priorscited by 22
  4. 2018RMPflow: A Computational Graph for Automatic Motion Policy Generationcited by 21
  5. 2022Learning Implicit Priors for Motion Optimizationcited by 20
  6. 2020Continuous Action Reinforcement Learning from a Mixture of Interpretable Expertscited by 17
  7. 2021Hierarchical Neural Dynamic Policiescited by 17
  8. 2022Safe Reinforcement Learning of Dynamic High-Dimensional Robotic Tasks: Navigation, Manipulation, Interactioncited by 17
  9. 2022Regularized Deep Signed Distance Fields for Reactive Motion Generationcited by 16
  10. 2020Merging Position and Orientation Motion Primitivescited by 12
  11. 2022Hierarchical Policy Blending as Inference for Reactive Robot Controlcited by 8
  12. 2020RMPflow: A Geometric Framework for Generation of Multi-Task Motion Policiescited by 7
  13. 2020Geometric Fabrics for the Acceleration-based Design of Robotic Motioncited by 7
  14. 2020Optimal Transport-based Coverage Control for Swarm Robot Systems: Generalization of the Voronoi Tessellation-based Methodcited by 7
  15. 2022Learning to Sequence and Blend Robot Skills via Differentiable Optimizationcited by 4
  16. 2021Adaptive Gaussian Process based Stochastic Trajectory Optimization for Motion Planningcited by 3
  17. 2022Fast Obstacle Avoidance Based on Real-Time Sensingcited by 3
  18. 2022Actor-Critic based Improper Reinforcement Learningcited by 3
  19. 2020Efficient, Decentralized, and Collaborative Multi-Robot Exploration using Optimal Transport Theorycited by 1

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 3.

4 papers.

  1. 2023Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Modelscited by 161
  2. 2024Efficient Diffusion Transformer Policies with Mixture of Expert Denoisers for Multitask Learningcited by 48
  3. 2024Acquiring Diverse Skills using Curriculum Reinforcement Learning with Mixture of Expertscited by 14
  4. 2024Dude: Dual Distribution-Aware Context Prompt Learning For Large Vision-Language Modelcited by 4