mlsanityData to 14 Oct 2026

Learning to Navigate in Mazes with Novel Layouts using Abstract Top-down Maps

arXiv 2412.12024 ↗First posted 16 Dec 2024cs.LG cs.AI cs.ROcited by 2

Not on the map. Why?

Abstract (arXiv)

Learning navigation capabilities in different environments has long been one of the major challenges in decision-making. In this work, we focus on zero-shot navigation ability using given abstract 2-D top-down maps. Like human navigation by reading a paper map, the agent reads the map as an image when navigating in a novel layout, after learning to navigate on a set of training maps. We propose a model-based reinforcement learning approach for this multi-task learning problem, where it jointly learns a hypermodel that takes top-down maps as input and predicts the weights of the transition network. We use the DeepMind Lab environment and customize layouts using generated maps. Our method can adapt better to novel environments in zero-shot and is more robust to noise.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 35 of its 45 references that are in the map. First posted on arXiv 16 Dec 2024.

Builds on

35 of 45 references.

References · 35

  1. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,353
  2. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  3. 2017Hindsight Experience Replaycited by 1,136
  4. 2016HyperNetworkscited by 1,027
  5. 2018Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Modelscited by 774
  6. 2017Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithmcited by 734
  7. 2018On Evaluation of Embodied Navigation Agentscited by 547
  8. 2016Reinforcement Learning with Unsupervised Auxiliary Taskscited by 404
  9. 2020Learning to Explore using Active Neural SLAMcited by 316
  10. 2019Continual learning with hypernetworkscited by 273
  11. 2018Distributed Prioritized Experience Replaycited by 269
  12. 2017Cognitive Mapping and Planning for Visual Navigationcited by 212
23 more
  1. 2016Learning to Navigate in Complex Environmentscited by 198
  2. 2015Learning Continuous Control Policies by Stochastic Value Gradientscited by 162
  3. 2019Benchmarking Model-Based Reinforcement Learningcited by 162
  4. 2016Value Iteration Networkscited by 160
  5. 2017Value Prediction Networkcited by 117
  6. 2019Learning Exploration Policies for Navigationcited by 116
  7. 2018Learning to Navigate in Cities Without a Mapcited by 104
  8. 2016Playing FPS Games with Deep Reinforcement Learningcited by 95
  9. 2017Neural Map: Structured Memory for Deep Reinforcement Learningcited by 90
  10. 2018Temporal Difference Models: Model-Free Deep RL for Model-Based Controlcited by 89
  11. 2016DeepMind Labcited by 74
  12. 2018Unsupervised Predictive Memory in a Goal-Directed Agentcited by 55
  13. 2019Mapping State Space using Landmarks for Universal Goal Reachingcited by 44
  14. 2020Continual Model-Based Reinforcement Learning with Hypernetworkscited by 44
  15. 2019ELF OpenGo: An Analysis and Open Reimplementation of AlphaZerocited by 39
  16. 2016Learning to Act by Predicting the Futurecited by 36
  17. 2019Generalization of Reinforcement Learners with Working and Episodic Memorycited by 36
  18. 2018Gated Path Planning Networkscited by 31
  19. 2020Discriminative Particle Filter Reinforcement Learning for Complex Partial Observationscited by 28
  20. 2019Differentiable Algorithm Networks for Composable Robot Learningcited by 24
  21. 2017Teaching a Machine to Read Maps with Deep Reinforcement Learningcited by 9
  22. 2022Integrating Symmetry into Differentiable Planning with Steerable Convolutionscited by 6
  23. 2022Scaling up and Stabilizing Differentiable Planning with Implicit Differentiationcited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026NavOne: One-Step Global Planning for Vision-Language Navigation on Top-Down Mapscited by 0
  2. 2026Neurosymbolic Reasoning with Incremental Knowledge for Sample Efficient Hierarchical Reinforcement Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Learning Spatial and Temporal Hierarchies: Hierarchical Active Inference for navigation in Multi-Room Maze Environments
  2. 2026A Reinforcement Learning-Based Model for Mapping and Goal-Directed Navigation Using Multiscale Place Fields
  3. 2025NavigateDiff: Visual Predictors are Zero-Shot Navigation Assistants
  4. 2025Bridging Deep Reinforcement Learning and Motion Planning for Model-Free Navigation in Cluttered Environments
  5. 2025Deep Reinforcement Learning Based Navigation with Macro Actions and Topological Maps
  6. 2023Inferring Hierarchical Structure in Multi-Room Maze Environments
  7. 2026Think before Go: Hierarchical Reasoning for Image-goal Navigation
  8. 2024TopV-Nav: Unlocking the Top-View Spatial Reasoning Potential of MLLM for Zero-shot Object Navigation