mlsanityData to 14 Oct 2026

Classical Planning in Deep Latent Space

arXiv 2107.00110 ↗First posted 30 Jun 2021cs.AI cs.LGcited by 22

Not on the map. Why?

Abstract (arXiv)

Current domain-independent, classical planners require symbolic models of the problem domain and instance as input, resulting in a knowledge acquisition bottleneck. Meanwhile, although deep learning has achieved significant success in many fields, the knowledge is encoded in a subsymbolic representation which is incompatible with symbolic systems such as planners. We propose Latplan, an unsupervised architecture combining deep learning and classical planning. Given only an unlabeled set of image pairs showing a subset of transitions allowed in the environment (training inputs), Latplan learns a complete propositional PDDL action model of the environment. Later, when a pair of images representing the initial and the goal states (planning inputs) is given, Latplan finds a plan to the goal state in a symbolic latent space and returns a visualized plan execution. We evaluate Latplan using image-based versions of 6 planning domains: 8-puzzle, 15-Puzzle, Blocksworld, Sokoban and Two variations of LightsOut.

Key facts. Cited by 22 papers in the map (data through 14 Oct 2026). Builds on 40 of its 159 references that are in the map. First posted on arXiv 30 Jun 2021.

Builds on

40 of 159 references.

References · 40

  1. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  2. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
  3. 2017Neural Discrete Representation Learningcited by 4,712
  4. 2013Auto-Encoding Variational Bayescited by 4,053
  5. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,611
  6. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  7. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,079
  8. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,353
  9. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,148
  10. 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 860
  11. 2019On the Variance of the Adaptive Learning Rate and Beyondcited by 772
  12. 2017Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithmcited by 734
28 more
  1. 2015Generating Sentences from a Continuous Spacecited by 612
  2. 2019Model-Based Reinforcement Learning for Ataricited by 502
  3. 2020Learning Invariant Representations for Reinforcement Learning without Reconstructioncited by 436
  4. 2014A* Samplingcited by 151
  5. 2014The LAMA Planner: Guiding Cost-Based Anytime Planning with Landmarkscited by 87
  6. 2018Bayesian Uncertainty Estimation for Batch Normalized Deep Networkscited by 70
  7. 2017Classical Planning in Deep Latent Space: Bridging the Subsymbolic-Symbolic Boundarycited by 58
  8. 2018Learning Plannable Representations with Causal InfoGANcited by 57
  9. 2019Learning Domain-Independent Planning Heuristics with Hypergraph Networkscited by 50
  10. 2020PDDLGym: Gym Environments from PDDL Problemscited by 45
  11. 2019Learning First-Order Symbolic Representations for Planning from the Structure of the State Spacecited by 38
  12. 2018Learning Finite State Representations of Recurrent Policy Networkscited by 37
  13. 2021Alchemy: A benchmark and analysis toolkit for meta-reinforcement learning agentscited by 36
  14. 2018DVAE++: Discrete Variational Autoencoders with Overlapping Transformationscited by 32
  15. 2015A Hierarchical Neural Autoencoder for Paragraphs and Documentscited by 26
  16. 2019Unsupervised Grounding of Plannable First-Order Logic Representation from Imagescited by 24
  17. 2020Understanding the Power and Limitations of Teaching with Imperfect Knowledgecited by 23
  18. 2021Learning First-Order Representations for Planning from Black-Box States: New Resultscited by 23
  19. 2018DVAE#: Discrete Variational Autoencoders with Relaxed Boltzmann Priorscited by 20
  20. 2019Learning STRIPS Action Models with Classical Planningcited by 14
  21. 2018Extracting Action Sequences from Texts Based on Deep Reinforcement Learningcited by 12
  22. 2017Active Exploration for Learning Symbolic Representationscited by 8
  23. 2018LSTM-Based Goal Recognition in Latent Spacecited by 8
  24. 2013An Unsupervised Approach for Automatic Activity Recognition based on Hidden Markov Model Regressioncited by 5
  25. 2019Towards Stable Symbol Grounding with Zero-Suppressed State AutoEncodercited by 5
  26. 2019Learning Action Models from Disordered and Noisy Plan Tracescited by 4
  27. 2018Photo-Realistic Blocksworld Datasetcited by 3
  28. 2019Neural-Symbolic Descriptive Action Model from Images: The Search for STRIPScited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 1; 2022: 1; 2023: 4; 2024: 2; 2025: 5; 2026: 9.

22 papers.

Sort:

By area:

  1. 2024Learning Lifted STRIPS Models from Action Traces Alone: A Simple, General, and Scalable Solutioncited by 8
  2. 2023Egocentric Planning for Scalable Embodied Task Achievementcited by 7
  3. 2024Learning Abstract World Model for Value-preserving Planning with Optionscited by 4
  4. 2025Neuro-Symbolic Imitation Learning: Discovering Symbolic Abstractions for Skill Learningcited by 4
  5. 2023Discovering Predictive Relational Object Symbols with Symbolic Attentive Layerscited by 3
  6. 2026Learning Lifted Action Models from Unsupervised Visual Tracescited by 3
  7. 2023A Review of Symbolic, Subsymbolic and Hybrid Methods for Sequential Decision Makingcited by 2
  8. 2025Learning Lifted Action Models From Traces of Incomplete Actions and Statescited by 2
  9. 2025From Next Token Prediction to (STRIPS) World Modelscited by 2
  10. 2021Is Policy Learning Overrated?: Width-Based Planning and Active Learning for Ataricited by 1
  11. 2025CoSPlan: Corrective Sequential Planning via Scene Graph Incremental Updatescited by 1
  12. 2026Discrete World Models via Regularizationcited by 1
10 more
  1. 2022Dr. Neurosymbolic, or: How I Learned to Stop Worrying and Accept Statisticscited by 0
  2. 2023Plausibility-Based Heuristics for Latent Space Classical Planningcited by 0
  3. 2025GRAML: Goal Recognition As Metric Learningcited by 0
  4. 2026The DeepXube Software Package for Solving Pathfinding Problems with Learned Heuristic Functions and Searchcited by 0
  5. 2026Differentiable Learning of Lifted Action Schemas for Classical Planningcited by 0
  6. 2026Learning Bilevel Policies over Symbolic World Models for Long-Horizon Planningcited by 0
  7. 2026Learning Lifted Action Models from Traces with Minimal Information About Actions and Statescited by 0
  8. 2026A Fully Differentiable Neuro-Soft-Symbolic Framework for Perceptual Task Planningcited by 0
  9. 2026Neurosymbolic Action Model Learning under Partial Observabilitycited by 0
  10. 2026Neurosymbolic Learning of Interpretable World Modelscited by 0