mlsanityData to 14 Oct 2026

Grounded Curriculum Learning

arXiv 2409.19816 ↗First posted 29 Sep 2024cs.RO cs.AIcited by 5

On the map: Deep reinforcement learning methods → Deep RL algorithms and representations

Abstract (arXiv)

The high cost of real-world data for robotics Reinforcement Learning (RL) leads to the wide usage of simulators. Despite extensive work on building better dynamics models for simulators to match with the real world, there is another, often-overlooked mismatch between simulations and the real world, namely the distribution of available training tasks. Such a mismatch is further exacerbated by existing curriculum learning techniques, which automatically vary the simulation task distribution without considering its relevance to the real world. Considering these challenges, we posit that curriculum learning for robotics RL needs to be grounded in real-world task distributions. To this end, we propose Grounded Curriculum Learning (GCL), which aligns the simulated task distribution in the curriculum with the real world, as well as explicitly considers what tasks have been given to the robot and how the robot has performed in the past. We validate GCL using the BARN dataset on complex navigation tasks, achieving a 6.8% and 6.5% higher success rate compared to a state-of-the-art CL method and a curriculum designed by human experts, respectively. These results show that GCL can enhance learning efficiency and navigation performance by grounding the simulation task distribution in the real world within an adaptive curriculum.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods. Builds on 43 of its 54 references that are in the map. First posted on arXiv 29 Sep 2024.

Builds on

43 of 54 references.

References · 43

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,498
  3. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 998
  4. 2018QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulationcited by 743
  5. 2017Robust Adversarial Reinforcement Learningcited by 451
  6. 2020Sim-to-Real Transfer in Deep Reinforcement Learning for Robotics: a Surveycited by 413
  7. 2020Curriculum Learning for Reinforcement Learning Domains: A Framework and Surveycited by 373
  8. 2019Challenges of Real-World Reinforcement Learningcited by 291
  9. 2020Emergent Complexity and Zero-shot Transfer via Unsupervised Environment Designcited by 275
  10. 2022Evolving Curricula with Regret-Based Environment Designcited by 173
  11. 2020Automatic Curriculum Learning through Value Disagreementcited by 87
  12. 2019A Survey of Deep Reinforcement Learning in Video Gamescited by 73
31 more
  1. 2018Composable Deep Reinforcement Learning for Robotic Manipulationcited by 72
  2. 2020Motion Planning and Control for Mobile Robot Navigation Using Machine Learning: a Surveycited by 58
  3. 2016From Perception to Decision: A Data-driven Approach to End-to-end Motion Planning for Autonomous Ground Robotscited by 44
  4. 2022Learning Model Predictive Controllers with Real-Time Attention for Real-World Navigationcited by 29
  5. 2021Learning Inverse Kinodynamics for Accurate High-Speed Off-Road Navigation on Unstructured Terraincited by 25
  6. 2022VI-IKD: High-Speed Accurate Off-Road Navigation using Learned Visual-Inertial Inverse Kinodynamicscited by 21
  7. 2020APPLD: Adaptive Planner Parameter Learning from Demonstrationcited by 20
  8. 2020Benchmarking Metric Ground Navigationcited by 20
  9. 2021APPL: Adaptive Planner Parameter Learningcited by 20
  10. 2022CLUTR: Curriculum Learning via Unsupervised Task Representation Learningcited by 20
  11. 2020APPLI: Adaptive Planner Parameter Learning From Interventionscited by 19
  12. 2022Benchmarking Reinforcement Learning Techniques for Autonomous Navigationcited by 19
  13. 2023Toward Wheeled Mobility on Vertically Challenging Terrain: Platforms, Datasets, and Algorithmscited by 19
  14. 2020APPLR: Adaptive Planner Parameter Learning from Reinforcementcited by 16
  15. 2021APPLE: Adaptive Planner Parameter Learning from Evaluative Feedbackcited by 16
  16. 2023Rethinking Social Robot Navigation: Leveraging the Best of Two Worldscited by 15
  17. 2024DTG : Diffusion-based Trajectory Generation for Mapless Global Navigationcited by 13
  18. 2020Toward Agile Maneuvers in Highly Constrained Spaces: Learning from Hallucinationcited by 11
  19. 2022High-Speed Accurate Robot Control using Learned Forward Kinodynamics and Non-linear Least Squares Optimizationcited by 11
  20. 2023MTG: Mapless Trajectory Generator with Traversability Coverage for Outdoor Navigationcited by 10
  21. 2021From Agile Ground to Aerial Navigation: Learning from Learned Hallucinationcited by 9
  22. 2020Agile Robot Navigation through Hallucinated Learning and Sober Deploymentcited by 7
  23. 2024Terrain-Attentive Learning for Efficient 6-DoF Kinodynamic Modeling on Vertically Challenging Terraincited by 7
  24. 2024Reinforcement Learning for Wheeled Mobility on Vertically Challenging Terraincited by 7
  25. 2023Motion Memory: Leveraging Past Experiences to Accelerate Future Motion Planningcited by 5
  26. 2021Team Orienteering Coverage Planning with Uncertain Rewardcited by 4
  27. 2023Team Coordination on Graphs with State-Dependent Edge Costcited by 4
  28. 2019On the use of Deep Autoencoders for Efficient Embedded Reinforcement Learningcited by 3
  29. 2022Learning Real-world Autonomous Navigation by Self-Supervised Environment Synthesiscited by 3
  30. 2024Scaling Team Coordination on Graphs with Reinforcement Learningcited by 3
  31. 2024Team Coordination on Graphs: Problem, Analysis, and Algorithmscited by 2

Built on by

Cited by year

Citing papers per year: 2024: 1; 2025: 4.

5 papers.

  1. 2024Adaptive Diffusion Terrain Generator for Autonomous Uneven Terrain Navigationcited by 6
  2. 2025Reward Training Wheels: Adaptive Auxiliary Rewards for Robotics Reinforcement Learningcited by 4
  3. 2025Adaptively Point-weighting Curriculum Learningcited by 1
  4. 2025TRACED: Transition-aware Regret Approximation with Co-learnability for Environment Designcited by 1
  5. 2025ADEPT: Adaptive Diffusion Environment for Policy Transfer Sim-to-Realcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Scaling Rough Terrain Locomotion with Automatic Curriculum Reinforcement Learning
  2. 2024Reverse Forward Curriculum Learning for Extreme Sample and Demonstration Efficiency in Reinforcement Learning
  3. 2022Teacher-student curriculum learning for reinforcement learning
  4. 2024Mobile Robots through Task-Based Human Instructions using Incremental Curriculum Learning
  5. 2025Causal-Paced Deep Reinforcement Learning
  6. 2025Probabilistic Curriculum Learning for Goal-Based Reinforcement Learning
  7. 2024CurricuLLM: Automatic Task Curricula Design for Learning Complex Robot Skills using Large Language Models
  8. 2026Active Curriculum Refinement for Reinforcement Learning