mlsanityData to 7 Oct 2026

Causal Navigation by Continuous-time Neural Networks

arXiv 2106.08314 ↗First posted 15 Jun 2021cs.LG cs.AI cs.NEcited by 23

On the map: Offline RL and imitation learning from data → Imitation learning methods

Abstract (arXiv)

Imitation learning enables high-fidelity, vision-based learning of policies within rich, photorealistic environments. However, such techniques often rely on traditional discrete-time neural models and face difficulties in generalizing to domain shifts by failing to account for the causal relationships between the agent and the environment. In this paper, we propose a theoretical and experimental framework for learning causal representations using continuous-time neural networks, specifically over their discrete-time counterparts. We evaluate our method in the context of visual-control learning of drones over a series of complex tasks, ranging from short- and long-term navigation, to chasing static and dynamic objects through photorealistic environments. Our results demonstrate that causal continuous-time deep models can perform robust navigation tasks, where advanced recurrent models fail. These models learn complex causal control representations directly from raw visual inputs and scale to solve a variety of tasks using imitation learning.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 50 of its 100 references that are in the map. First posted on arXiv 15 Jun 2021.

Builds on

50 of 100 references.

Closest ancestors · 5

Same direction, newest first

  1. 2020Fighting Copycat Agents in Behavioral Cloning from Observation Histories28 Oct
  2. 2019Adversarial Imitation Learning from Incomplete Demonstrations29 May
  3. 2019Causal Confusion in Imitation Learning28 May
  4. 2019Imitation Learning from Imperfect Demonstration27 Jan
  5. 2018Deep Imitative Models for Flexible Inference, Planning, and Control15 Oct

Other references · 45

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2018Neural Ordinary Differential Equationscited by 3,733
  4. 2014Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modelingcited by 2,341
  5. 2016Generative Adversarial Imitation Learningcited by 1,576
  6. 2018An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modelingcited by 1,447
  7. 2019Neural Spline Flowscited by 590
  8. 2018FFJORD: Free-form Continuous Dynamics for Scalable Reversible Generative Modelscited by 584
  9. 2020Learning Quadrupedal Locomotion over Challenging Terraincited by 494
  10. 2019Augmented Neural ODEscited by 396
  11. 2019Learning by Cheatingcited by 361
  12. 2020Learning to Explore using Active Neural SLAMcited by 308
33 more
  1. 2016On the Expressive Power of Deep Neural Networkscited by 302
  2. 2020Algorithmic Recourse: from Counterfactual Explanations to Interventionscited by 250
  3. 2019Causality for Machine Learningcited by 242
  4. 2017One-Shot Imitation Learningcited by 232
  5. 2017Cognitive Mapping and Planning for Visual Navigationcited by 209
  6. 2018Semi-parametric Topological Memory for Navigationcited by 203
  7. 2020Neural Topological SLAM for Visual Navigationcited by 193
  8. 2019Latent ODEs for Irregularly-Sampled Time Seriescited by 182
  9. 2020Liquid Time-constant Networkscited by 160
  10. 2020Dissecting Neural ODEscited by 150
  11. 2019Neural Jump Stochastic Differential Equationscited by 147
  12. 2020Can Autonomous Vehicles Identify, Recover From, and Adapt to Distribution Shifts?cited by 139
  13. 2020Learning to Control PDEs with Differentiable Physicscited by 131
  14. 2019Learning Exploration Policies for Navigationcited by 113
  15. 2018Visual Representations for Semantic Target Driven Navigationcited by 89
  16. 2020Lipschitz Recurrent Neural Networkscited by 88
  17. 2019ANODE: Unconditionally Accurate Memory-Efficient Gradients for Neural ODEscited by 83
  18. 2019On Robustness of Neural Ordinary Differential Equationscited by 83
  19. 2020Learning Long-Term Dependencies in Irregularly-Sampled Time Seriescited by 82
  20. 2020Coupled Oscillatory Recurrent Neural Network (coRNN): An accurate and (gradient) stable architecture for learning long time dependenciescited by 81
  21. 2018Conditional Affordance Learning for Driving in Urban Environmentscited by 70
  22. 2020Diverse and Admissible Trajectory Forecasting through Multimodal Context Understandingcited by 68
  23. 2020Deep Drone Acrobaticscited by 66
  24. 2014From Ordinary Differential Equations to Structural Causal Models: the deterministic casecited by 41
  25. 2016From Deterministic ODEs to Dynamic Structural Causal Modelscited by 32
  26. 2018One-Shot Hierarchical Imitation Learning of Compound Visuomotor Taskscited by 32
  27. 2017Discrete Event, Continuous Time RNNscited by 31
  28. 2019SNODE: Spectral Discretization of Neural ODEs for System Identificationcited by 26
  29. 2020On The Verification of Neural ODEs with Stochastic Guaranteescited by 24
  30. 2021Adversarial Training is Not Ready for Robot Learningcited by 20
  31. 2020Causal structure learning from time series: Large regression coefficients may predict causal links better in practice than small p-valuescited by 18
  32. 2020AirSim Drone Racing Labcited by 15
  33. 2017Unifying Map and Landmark Based Representations for Visual Navigationcited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 7; 2023: 7; 2024: 3; 2025: 1; 2026: 1.

23 papers.

Sort:

By area:

  1. 2022Liquid Structural State-Space Modelscited by 134
  2. 2021Closed-form Continuous-time Neural Modelscited by 65
  3. 2023Dataset Distillation with Convexified Implicit Gradientscited by 46
  4. 2024Trajectory Flow Matching with Applications to Clinical Time Series Modelingcited by 21
  5. 2022Evolution of Neural Tangent Kernels under Benign and Adversarial Trainingcited by 14
  6. 2021Sparse Flows: Pruning Continuous-depth Modelscited by 13
  7. 2021GoTube: Scalable Stochastic Verification of Continuous-Depth Modelscited by 12
  8. 2024Gaussian Splatting to Real World Flight Navigation Transfer with Liquid Networkscited by 12
  9. 2021BarrierNet: A Safety-Guaranteed Layer for Neural Networkscited by 11
  10. 2022On the Forward Invariance of Neural ODEscited by 10
  11. 2023A Survey of Methods, Challenges and Perspectives in Causalitycited by 6
  12. 2022Are All Vision Models Created Equal? A Study of the Open-Loop to Closed-Loop Causality Gapcited by 5
11 more
  1. 2023Learning with Chemical versus Electrical Synapses -- Does it Make a Difference?cited by 5
  2. 2022Entangled Residual Mappingscited by 4
  3. 2024Liquid Fourier Latent Dynamics Networks for fast GPU-based numerical simulations in computational cardiologycited by 3
  4. 2022Towards Cooperative Flight Control Using Visual-Attentioncited by 2
  5. 2023TLNets: Transformation Learning Networks for long-range time-series predictioncited by 2
  6. 2023Leveraging Low-Rank and Sparse Recurrent Connectivity for Robust Closed-Loop Controlcited by 2
  7. 2022Interpreting Neural Policies with Disentangled Tree Representationscited by 1
  8. 2023Infrastructure-based End-to-End Learning and Prevention of Driver Failurecited by 0
  9. 2023Robustness Analysis of Continuous-Depth Models with Lagrangian Techniquescited by 0
  10. 2025The power of dynamic causality in observer-based design for soft sensor applicationscited by 0
  11. 2026Real-time fall detection based on vision for low-power edge platformscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2019Learning Visuomotor Policies for Aerial Navigation Using Cross-Modal Representations
  2. 2023Interpretable Imitation Learning with Dynamic Causal Relations
  3. 2026DreamFly: Causal Memory and Receding-Horizon Diffusion Planning for Aerial Vision-Language Navigation
  4. 2026Causal World Modeling for Robot Control
  5. 2026Scalable Causal Imitation Learning
  6. 2021Robust Navigation for Racing Drones based on Imitation Learning and Modularization
  7. 2026TRACE: Trajectory Recovery for Continuous Mechanism Evolution in Causal Representation Learning
  8. 2020Approximate Inverse Reinforcement Learning from Vision-based Imitation Learning