mlsanityData to 7 Oct 2026

Learning Neuro-Symbolic Skills for Bilevel Planning

arXiv 2206.10680 ↗First posted 21 Jun 2022cs.RO cs.AI cs.LGcited by 55

Not on the map. Why?

Abstract (arXiv)

Decision-making is challenging in robotics environments with continuous object-centric states, continuous actions, long horizons, and sparse feedback. Hierarchical approaches, such as task and motion planning (TAMP), address these challenges by decomposing decision-making into two or more levels of abstraction. In a setting where demonstrations and symbolic predicates are given, prior work has shown how to learn symbolic operators and neural samplers for TAMP with manually designed parameterized policies. Our main contribution is a method for learning parameterized polices in combination with operators and samplers. These components are packaged into modular neuro-symbolic skills and sequenced together with search-then-sample TAMP to solve new tasks. In experiments in four robotics domains, we show that our approach -- bilevel planning with neuro-symbolic skills -- can solve a wide range of tasks with varying initial states, goals, and objects, outperforming six baselines and ablations. Video: https://youtu.be/PbFZP8rPuGg Code: https://tinyurl.com/skill-learning

Key facts. Cited by 55 papers in the map (data through 7 Oct 2026). Builds on 42 of its 105 references that are in the map. First posted on arXiv 21 Jun 2022.

Builds on

42 of 105 references.

References · 42

  1. 2022Do As I Can, Not As I Say: Grounding Language in Robotic Affordancescited by 2,302
  2. 2017Hindsight Experience Replaycited by 1,096
  3. 2022Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agentscited by 1,029
  4. 2018Diversity is All You Need: Learning Skills without a Reward Functioncited by 734
  5. 2021Implicit Behavioral Cloningcited by 442
  6. 2018Data-Efficient Hierarchical Reinforcement Learningcited by 436
  7. 2016The Option-Critic Architecturecited by 433
  8. 2017FeUdal Networks for Hierarchical Reinforcement Learningcited by 328
  9. 2020Integrated Task and Motion Planningcited by 281
  10. 2016Variational Intrinsic Controlcited by 262
  11. 2020Accelerating Reinforcement Learning with Learned Skill Priorscited by 220
  12. 2022Pre-Trained Language Models for Interactive Decision-Makingcited by 220
30 more
  1. 2020OPAL: Offline Primitive Discovery for Accelerating Offline Reinforcement Learningcited by 164
  2. 2020Parrot: Data-Driven Behavioral Priors for Reinforcement Learningcited by 136
  3. 2020Learning to Generalize Across Long-Horizon Tasks from Human Demonstrationscited by 114
  4. 2021Skill Induction and Planning with Latent Languagecited by 99
  5. 2017Meta Learning Shared Hierarchiescited by 93
  6. 2021Bottom-Up Skill Discovery from Unsegmented Demonstrations for Long-Horizon Robot Manipulationcited by 82
  7. 2018SDRL: Interpretable and Data-efficient Deep Reinforcement Learning Leveraging Symbolic Planningcited by 77
  8. 2017Near Optimal Behavior via Approximate State Abstractioncited by 63
  9. 2021Learning Symbolic Operators for Task and Motion Planningcited by 57
  10. 2018PEORL: Integrating Symbolic Planning and Hierarchical Reinforcement Learning for Robust Decision-Makingcited by 51
  11. 2017Multi-Level Discovery of Deep Optionscited by 50
  12. 2020Learning compositional models of robot skills for task and motion planningcited by 47
  13. 2020Deep Visual Reasoning: Learning to Predict Action Sequences for Task and Motion Planning from an Initial Scene Imagecited by 45
  14. 2021Learning Neuro-Symbolic Relational Transition Models for Bilevel Planningcited by 43
  15. 2018Learning Abstract Optionscited by 31
  16. 2020Understanding the Power and Limitations of Teaching with Imperfect Knowledgecited by 23
  17. 2020SPOTTER: Extending Symbolic Planning Operators through Targeted Reinforcement Learningcited by 23
  18. 2021SKID RAW: Skill Discovery from Raw Trajectoriescited by 22
  19. 2022Leveraging Approximate Symbolic Models for Reinforcement Learning via Skill Diversitycited by 22
  20. 2022Generalizable Task Planning through Representation Pretrainingcited by 19
  21. 2021Discovering State and Action Abstractions for Generalized Task and Motion Planningcited by 17
  22. 2019Generalized Lazy Search for Robot Motion Planning: Interleaving Search and Edge Evaluation via Event-based Togglescited by 16
  23. 2018Learning Quickly to Plan Quickly Using Modular Meta-Learningcited by 15
  24. 2021Extended Tree Search for Robot Task and Motion Planningcited by 14
  25. 2017Independently Controllable Featurescited by 11
  26. 2019Learning Portable Representations for High-Level Planningcited by 11
  27. 2021Discovering User-Interpretable Capabilities of Black-Box Planning Agentscited by 10
  28. 2022Temporal Logic Imitation: Learning Plan-Satisficing Motion Policies from Demonstrationscited by 10
  29. 2016Do What I Want, Not What I Did: Imitation of Skills by Planning Sequences of Actionscited by 3
  30. 2019Learning Deep Parameterized Skills from Demonstration for Re-targetable Visuomotor Controlcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 5; 2023: 10; 2024: 16; 2025: 15; 2026: 9.

55 papers.

Sort:

By area:

  1. 2023Text2Motion: From Natural Language Instructions to Feasible Planscited by 223
  2. 2024LoTa-Bench: Benchmarking Language-oriented Task Planners for Embodied Agentscited by 51
  3. 2024VisualPredicator: Learning Abstract World Models with Neuro-Symbolic Predicates for Robot Planningcited by 37
  4. 2025Neuro-Symbolic AI in 2024: A Systematic Reviewcited by 36
  5. 2022Towards Data-and Knowledge-Driven Artificial Intelligence: A Survey on Neuro-Symbolic Computingcited by 34
  6. 2022LEAGUE: Guided Skill Learning and Abstraction for Long-Horizon Manipulationcited by 32
  7. 2023Logic-induced Diagnostic Reasoning for Semi-supervised Semantic Segmentationcited by 29
  8. 2023Human-in-the-Loop Task and Motion Planning for Imitation Learningcited by 28
  9. 2023Neurosymbolic Reinforcement Learning and Planning: A Surveycited by 24
  10. 2024Practice Makes Perfect: Planning to Learn Skill Parameter Policiescited by 22
  11. 2023DiMSam: Diffusion Models as Samplers for Task and Motion Planning under Partial Observabilitycited by 21
  12. 2024A Survey of Optimization-based Task and Motion Planning: From Classical To Learning Approachescited by 19
38 more
  1. 2022Learning Efficient Abstract Planning Models that Choose What to Predictcited by 17
  2. 2023Embodied Lifelong Learning for Task and Motion Planningcited by 17
  3. 2024Composing Pre-Trained Object-Centric Representations for Robotics From "What" and "Where" Foundation Modelscited by 16
  4. 2025Synthesizing world models for bilevel planningcited by 13
  5. 2026LiLo-VLA: Compositional Long-Horizon Manipulation via Linked Object-Centric Policiescited by 12
  6. 2024From Pixels to Predicates: Learning Symbolic World Models via Pretrained Vision-Language Modelscited by 11
  7. 2024Neuro-Symbolic AI: Explainability, Challenges, and Future Trendscited by 10
  8. 2023Embodied Active Learning of Relational State Abstractions for Bilevel Planningcited by 8
  9. 2024LogiCity: Advancing Neuro-Symbolic AI with Abstract Urban Simulationcited by 8
  10. 2025ExoPredicator: Learning Abstract Models of Dynamic Worlds for Robot Planningcited by 8
  11. 2022Relative Behavioral Attributes: Filling the Gap between Symbolic Goal Specification and Reward Learning from Human Preferencescited by 7
  12. 2023Robot Task Planning Based on Large Language Model Representing Knowledge with Directed Graph Structurescited by 7
  13. 2024Task and Motion Planning for Execution in the Realcited by 6
  14. 2024MAPLE: A Framework for Active Preference Learning Guided by Large Language Modelscited by 6
  15. 2025Neuro-Symbolic Imitation Learning: Discovering Symbolic Abstractions for Skill Learningcited by 4
  16. 2025Prime the search: Using large language models for guiding geometric task and motion planning by warm-starting tree searchcited by 4
  17. 2025A Systematic Study of Large Language Models for Task and Motion Planning With PDDLStreamcited by 4
  18. 2022Learning Temporally Extended Skills in Continuous Domains as Symbolic Actions for Planningcited by 3
  19. 2023Discovering Predictive Relational Object Symbols with Symbolic Attentive Layerscited by 3
  20. 2025Curiosity-Driven Imagination: Discovering Plan Operators and Learning Associated Policies for Open-World Adaptationcited by 3
  21. 2025LodeStar: Long-horizon Dexterity via Synthetic Data Augmentation from Human Demonstrationscited by 3
  22. 2025SLAP: Shortcut Learning for Abstract Planningcited by 3
  23. 2025MOSAIC: Skill-Centric Manipulation Planning with Physics Simulationcited by 2
  24. 2025Few-Shot Neuro-Symbolic Imitation Learning for Long-Horizon Planning and Actingcited by 2
  25. 2024SYNAPSE: SYmbolic Neural-Aided Preference Synthesis Enginecited by 1
  26. 2024Learning to Bridge the Gap: Efficient Novelty Recovery with Planning and Reinforcement Learningcited by 1
  27. 2024Neuro-Symbolic Skill Discovery for Conditional Multi-Level Planningcited by 1
  28. 2024Learning for Long-Horizon Planning via Neuro-Symbolic Abductive Imitationcited by 1
  29. 2025Situationally-Aware Dynamics Learningcited by 1
  30. 2025Finding 3D Scene Analogies with Multimodal Foundation Modelscited by 1
  31. 2026Novelty Adaptation Through Hybrid Large Language Model (LLM)-Symbolic Planning and LLM-guided Reinforcement Learningcited by 1
  32. 2023Learning Top-k Subtask Planning Tree based on Discriminative Representation Pre-training for Decision Makingcited by 0
  33. 2024NSP: A Neuro-Symbolic Natural Language Navigational Plannercited by 0
  34. 2024Bridging the Gap: Representation Spaces in Neuro-Symbolic AIcited by 0
  35. 2025Task and Skill Planning: Hierarchical Robot Planning with Black-Box Skillscited by 0
  36. 2025Using Temperature Sampling to Effectively Train Robot Learning Policies on Imbalanced Datasetscited by 0
  37. 2026Continual learning and refinement of causal models through dynamic predicate inventioncited by 0
  38. 2026Build on Priors: Vision--Language--Guided Neuro-Symbolic Imitation Learning for Data-Efficient Real-World Robot Manipulationcited by 0

50 most cited of 55 All 55 (JSON)