mlsanityData to 14 Oct 2026

Graph augmented Deep Reinforcement Learning in the GameRLand3D environment

arXiv 2112.11731 ↗First posted 22 Dec 2021cs.LG cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

We address planning and navigation in challenging 3D video games featuring maps with disconnected regions reachable by agents using special actions. In this setting, classical symbolic planners are not applicable or difficult to adapt. We introduce a hybrid technique combining a low level policy trained with reinforcement learning and a graph based high level classical planner. In addition to providing human-interpretable paths, the approach improves the generalization performance of an end-to-end approach in unseen maps, where it achieves a 20% absolute increase in success rate over a recurrent end-to-end agent on a point to point navigation task in yet unseen large-scale maps of size 1km x 1km. In an in-depth experimental study, we quantify the limitations of end-to-end Deep RL approaches in vast environments and we also introduce "GameRLand3D", a new benchmark and soon to be released environment can generate complex procedural 3D maps for navigation tasks.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 54 of its 83 references that are in the map. First posted on arXiv 22 Dec 2021.

Builds on

54 of 83 references.

References · 54

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  3. 2015Continuous control with deep reinforcement learningcited by 3,407
  4. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  5. 2016Asynchronous Methods for Deep Reinforcement Learningcited by 2,258
  6. 2020Conservative Q-Learning for Offline Reinforcement Learningcited by 1,809
  7. 2016OpenAI Gymcited by 1,736
  8. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,240
  9. 2017Curiosity-driven Exploration by Self-supervised Predictioncited by 1,194
  10. 2019Habitat: A Platform for Embodied AI Researchcited by 1,191
  11. 2020Mastering Atari with Discrete World Modelscited by 1,007
  12. 2018IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architecturescited by 883
38 more
  1. 2017AI2-THOR: An Interactive 3D Environment for Visual AIcited by 869
  2. 2017Rainbow: Combining Improvements in Deep Reinforcement Learningcited by 786
  3. 2014Neural Turing Machinescited by 644
  4. 2017Evolution Strategies as a Scalable Alternative to Reinforcement Learningcited by 574
  5. 2018On Evaluation of Embodied Navigation Agentscited by 547
  6. 2021Habitat 2.0: Training Home Assistants to Rearrange their Habitatcited by 511
  7. 2018Gibson Env: Real-World Perception for Embodied Agentscited by 481
  8. 2020Object Goal Navigation using Goal-Oriented Semantic Explorationcited by 429
  9. 2016Reinforcement Learning with Unsupervised Auxiliary Taskscited by 404
  10. 2016Learning to reinforcement learncited by 377
  11. 2019DD-PPO: Learning Near-Perfect PointGoal Navigators from 2.5 Billion Framescited by 362
  12. 2020Learning to Explore using Active Neural SLAMcited by 316
  13. 2018Distributed Prioritized Experience Replaycited by 269
  14. 2018Human-level performance in first-person multiplayer games with population-based deep reinforcement learningcited by 264
  15. 2018Unity: A General Platform for Intelligent Agentscited by 236
  16. 2017Cognitive Mapping and Planning for Visual Navigationcited by 212
  17. 2018Semi-parametric Topological Memory for Navigationcited by 209
  18. 2016Learning to Navigate in Complex Environmentscited by 198
  19. 2020Neural Topological SLAM for Visual Navigationcited by 195
  20. 2019Search on the Replay Buffer: Bridging Planning and Reinforcement Learningcited by 194
  21. 2019OpenSpiel: A Framework for Reinforcement Learning in Gamescited by 186
  22. 2019Soft Actor-Critic for Discrete Action Settingscited by 174
  23. 2020Acme: A Research Framework for Distributed Reinforcement Learningcited by 146
  24. 2018Learning to Navigate in Cities Without a Mapcited by 104
  25. 2016Playing FPS Games with Deep Reinforcement Learningcited by 95
  26. 2020MultiON: Benchmarking Semantic Map Memory using Multi-Object Navigationcited by 95
  27. 2020Sample Factory: Egocentric 3D Control from Pixels at 100000 FPS with Asynchronous Reinforcement Learningcited by 93
  28. 2017Neural Map: Structured Memory for Deep Reinforcement Learningcited by 90
  29. 2016DeepMind Labcited by 74
  30. 2018Unsupervised Predictive Memory in a Goal-Directed Agentcited by 55
  31. 2015Gated Feedback Recurrent Neural Networkscited by 54
  32. 2018Emergence of grid-like representations by training recurrent neural networks to perform spatial localizationcited by 52
  33. 2018ViZDoom Competitions: Playing Doom from Pixelscited by 51
  34. 2019Making Efficient Use of Demonstrations to Solve Hard Exploration Problemscited by 45
  35. 2020Semantic Curiosity for Active Visual Learningcited by 45
  36. 2018Accelerated Methods for Deep Reinforcement Learningcited by 40
  37. 2021Adversarial Reinforcement Learning for Procedural Content Generationcited by 36
  38. 2020Learning to plan with uncertain topological mapscited by 29

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 2; 2025: 1.

5 papers.

  1. 2023Improving Intrinsic Exploration by Creating Stationary Objectivescited by 7
  2. 2024Learning to navigate efficiently and precisely in real environmentscited by 5
  3. 2025Offline Goal-Conditioned Reinforcement Learning with Projective Quasimetric Planningcited by 5
  4. 2023Learning Computational Efficient Bots with Costly Featurescited by 2
  5. 2024Effective Reward Specification in Deep Reinforcement Learningcited by 0