mlsanityData to 13 Oct 2026

Scaling up and Stabilizing Differentiable Planning with Implicit Differentiation

arXiv 2210.13542 ↗First posted 24 Oct 2022cs.LG cs.AI cs.ROcited by 6

Not on the map. Why?

Abstract (arXiv)

Differentiable planning promises end-to-end differentiability and adaptivity. However, an issue prevents it from scaling up to larger-scale problems: they need to differentiate through forward iteration layers to compute gradients, which couples forward computation and backpropagation, and needs to balance forward planner performance and computational cost of the backward pass. To alleviate this issue, we propose to differentiate through the Bellman fixed-point equation to decouple forward and backward passes for Value Iteration Network and its variants, which enables constant backward cost (in planning horizon) and flexible forward budget and helps scale up to large tasks. We study the convergence stability, scalability, and efficiency of the proposed implicit version of VIN and its variants and demonstrate their superiorities on a range of planning tasks: 2D navigation, visual navigation, and 2-DOF manipulation in configuration space and workspace.

Key facts. Cited by 6 papers in the map (data through 13 Oct 2026). Builds on 31 of its 43 references that are in the map. First posted on arXiv 24 Oct 2022.

Builds on

31 of 43 references.

References · 31

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,950
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,656
  3. 2018Neural Ordinary Differential Equationscited by 3,750
  4. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,620
  5. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,334
  6. 2019Deep Equilibrium Modelscited by 606
  7. 2017OptNet: Differentiable Optimization as a Layer in Neural Networkscited by 592
  8. 2019General E(2)-Equivariant Steerable CNNscited by 442
  9. 2016Value Iteration Networkscited by 158
  10. 2020Multiscale Deep Equilibrium Modelscited by 150
  11. 2017Imagination-Augmented Agents for Deep Reinforcement Learningcited by 138
  12. 2019Implicit Deep Learningcited by 124
19 more
  1. 2017Value Prediction Networkcited by 117
  2. 2020Monotone operator equilibrium networkscited by 100
  3. 2019Exploring Model-based Planning with Policy Networkscited by 90
  4. 2020The Value Equivalence Principle for Model-Based Reinforcement Learningcited by 90
  5. 2018Temporal Difference Models: Model-Free Deep RL for Model-Based Controlcited by 89
  6. 2021Stabilizing Equilibrium Models by Jacobian Regularizationcited by 71
  7. 2019An investigation of model-free planningcited by 64
  8. 2020Model-Augmented Actor-Critic: Backpropagating through Pathscited by 61
  9. 2017QMDP-Net: Deep Learning for Planning under Partial Observabilitycited by 51
  10. 2021Control-Oriented Model-Based Reinforcement Learning with Implicit Differentiationcited by 35
  11. 2021Proper Value Equivalencecited by 33
  12. 2018Gated Path Planning Networkscited by 31
  13. 2018Universal Planning Networkscited by 29
  14. 2019The Differentiable Cross-Entropy Methodcited by 27
  15. 2021Differentiable Spatial Planning using Transformerscited by 25
  16. 2021Neural Algorithmic Reasoners are Implicit Plannerscited by 21
  17. 2017Generalized Value Iteration Networks: Life Beyond Latticescited by 16
  18. 2019Value Iteration Networks on Multiple Levels of Abstractioncited by 11
  19. 2022Integrating Symmetry into Differentiable Planning with Steerable Convolutionscited by 6

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2023: 2; 2024: 3; 2025: 0; 2026: 1.

6 papers.

  1. 2023E(2)-Equivariant Graph Planning for Navigationcited by 6
  2. 2024Equivariant Action Sampling for Reinforcement Learning and Planningcited by 4
  3. 2024Scaling Value Iteration Networks to 5000 Layers for Extreme Long-Term Planningcited by 3
  4. 2024Learning to Navigate in Mazes with Novel Layouts using Abstract Top-down Mapscited by 2
  5. 2023Can Euclidean Symmetry be Leveraged in Reinforcement Learning and Planning?cited by 1
  6. 2026Compact Bellman-Grounded Cognitive Maps for Cost-Aware Navigationcited by 0