mlsanityData to 7 Oct 2026

Reverse Curriculum Generation for Reinforcement Learning

arXiv 1707.05300 ↗First posted 17 Jul 2017cs.AI cs.LG cs.NEcited by 185

Not on the map. Why?

Abstract (arXiv)

Many relevant tasks require an agent to reach a certain state, or to manipulate objects into a desired configuration. For example, we might want a robot to align and assemble a gear onto an axle or insert and turn a key in a lock. These goal-oriented tasks present a considerable challenge for reinforcement learning, since their natural reward function is sparse and prohibitive amounts of exploration are required to reach the goal and receive some learning signal. Past approaches tackle these problems by exploiting expert demonstrations or by manually designing a task-specific reward shaping function to guide the learning agent. Instead, we propose a method to learn these tasks without requiring any prior knowledge other than obtaining a single state in which the task is achieved. The robot is trained in reverse, gradually learning to reach the goal from a set of start states increasingly far from the goal. Our method automatically generates a curriculum of start states that adapts to the agent's performance, leading to efficient training on goal-oriented tasks. We demonstrate our approach on difficult simulated navigation and fine-grained manipulation problems, not solvable by state-of-the-art reinforcement learning methods.

Key facts. Cited by 185 papers in the map (data through 7 Oct 2026). First posted on arXiv 17 Jul 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2019: 2; 2020: 3; 2021: 25; 2022: 33; 2023: 30; 2024: 22; 2025: 27; 2026: 43.

185 papers.

Sort:

By area:

  1. 2021Curriculum Learning: A Surveycited by 257
  2. 2022Evolving Curricula with Regret-Based Environment Designcited by 166
  3. 2019Goal-conditioned Imitation Learningcited by 159
  4. 2022Exploration in Deep Reinforcement Learning: A Surveycited by 158
  5. 2022Curriculum Temperature for Knowledge Distillationcited by 110
  6. 2025Self-Evolving Curriculum for LLM Reasoningcited by 107
  7. 2022Jump-Start Reinforcement Learningcited by 102
  8. 2024StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedbackcited by 74
  9. 2021Asymmetric self-play for automatic goal discovery in robotic manipulationcited by 71
  10. 2024Training Large Language Models for Reasoning through Reverse Curriculum Reinforcement Learningcited by 63
  11. 2023How Simulation Helps Autonomous Driving:A Survey of Sim2real, Digital Twins, and Parallel Intelligencecited by 55
  12. 2022Exploration via Elliptical Episodic Bonusescited by 51
38 more
  1. 2022Improving Intrinsic Exploration with Language Abstractionscited by 49
  2. 2023Learning to Generate Better Than Your LLMcited by 49
  3. 2023Towards A Unified Agent with Foundation Modelscited by 46
  4. 2024Adaptive Data Optimization: Dynamic Sample Selection with Scaling Lawscited by 46
  5. 2021Heuristic-Guided Reinforcement Learningcited by 41
  6. 2021Autonomous Reinforcement Learning via Subgoal Curriculacited by 32
  7. 2022LEAGUE: Guided Skill Learning and Abstraction for Long-Horizon Manipulationcited by 32
  8. 2025RollPacker: Mitigating Long-Tail Rollouts for Fast, Synchronous RL Post-Trainingcited by 31
  9. 2025Training a Generally Curious Agentcited by 29
  10. 2022Learning to Solve Travelling Salesman Problem with Hardness-adaptive Curriculumcited by 27
  11. 2022Learning Dexterous Manipulation from Exemplar Object Trajectories and Pre-Graspscited by 27
  12. 2023ASP: Learn a Universal Neural Solver!cited by 27
  13. 2021Self-Paced Context Evaluation for Contextual Reinforcement Learningcited by 25
  14. 2021Variational Automatic Curriculum Learning for Sparse-Reward Cooperative Multi-Agent Problemscited by 25
  15. 2023Provable Advantage of Curriculum Learning on Parity Targets with Mixed Inputscited by 22
  16. 2021A Probabilistic Interpretation of Self-Paced Learning with Applications to Reinforcement Learningcited by 21
  17. 2021GridToPix: Training Embodied Agents with Minimal Supervisioncited by 20
  18. 2024A Single Goal is All You Need: Skills and Exploration Emerge from Contrastive RL without Rewards, Demonstrations, or Subgoalscited by 20
  19. 2023Proximal Curriculum for Reinforcement Learning Agentscited by 19
  20. 2024CurricuLLM: Automatic Task Curricula Design for Learning Complex Robot Skills using Large Language Modelscited by 19
  21. 2022Topological Experience Replaycited by 17
  22. 2023One ACT Play: Single Demonstration Behavior Cloning with Action Chunking Transformerscited by 17
  23. 2022Understanding the Complexity Gains of Single-Task RL with a Curriculumcited by 16
  24. 2024Boost Your Human Image Generation Model via Direct Preference Optimizationcited by 16
  25. 2021C-Planning: An Automatic Curriculum for Learning Goal-Reaching Taskscited by 15
  26. 2023Domain Randomization via Entropy Maximizationcited by 15
  27. 2025RL for Reasoning by Adaptively Revealing Rationalescited by 15
  28. 2024Acquiring Diverse Skills using Curriculum Reinforcement Learning with Mixture of Expertscited by 14
  29. 2024Eurekaverse: Environment Curriculum Generation via Large Language Modelscited by 14
  30. 2023Reward-Free Curricula for Training Robust World Modelscited by 13
  31. 2024Reverse Forward Curriculum Learning for Extreme Sample and Demonstration Efficiency in Reinforcement Learningcited by 12
  32. 2021Hierarchies of Planning and Reinforcement Learning for Robot Navigationcited by 11
  33. 2022Automating Reinforcement Learning with Example-based Resetscited by 11
  34. 2024Can LLM be a Good Path Planner based on Prompt Engineering? Mitigating the Hallucination for Path Planningcited by 11
  35. 2020Episodic Self-Imitation Learning with Hindsightcited by 10
  36. 2022Robust Reinforcement Learning via Genetic Curriculumcited by 10
  37. 2023Imitation Learning from Observation with Automatic Discount Schedulingcited by 10
  38. 2021Coarse-to-Fine Curriculum Learningcited by 9

50 most cited of 185 All 185 (JSON)