Research map / Probabilistic and scientific modeling / Dynamical systems
Neural Control: Adjoint Learning Through Equilibrium Constraints
Dezhong Tong, Jiawen Wang, Hengyi Zhou, Yinlong Shen, Xiaonan Huang, M. Khalid Jawed
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add receding-horizon continuation to adjoint control (ablation) | Neural Control · Task 3 multistable shape formation | best curvature loss: 8.4×10−3 -> 3.8×10−8 | Improved |
| Add receding-horizon continuation to adjoint control (ablation) | Neural Control · Task 2 trajectory tracking | best loss: 4.3×10−4 -> 3.6×10−8 | Improved |
| Add receding-horizon continuation to adjoint control (ablation) | Neural Control · Task 1 point-to-point | best loss: 3.2×10−7 -> 2.3×10−7 | No clear effect |
| Add receding-horizon continuation to adjoint control (ablation) | Neural Control · Task 3 | wall-clock time (s): 935.2 -> 50.9 | Improved |
| Use adjoint gradients instead of SPSA | Neural Control · Task 2 trajectory tracking | best loss: 5.4×10−3 -> 4.3×10−4 (Adjoint-only) | Improved |
| Use Adjoint+RHC instead of iCEM | Neural Control · Task 3 | wall-clock time (s): 9907.3 -> 50.9 | Improved |
| Use Adjoint+RHC instead of SPSA | Neural Control · Task 1 point-to-point | wall-clock time (s): 727.3 -> 16.1 | Improved |
| Use SPSA instead of Adjoint-only on Task 3 loss | SPSA · Task 3 | best loss: 8.4×10−3 -> 2.8×10−5 | Improved |
Closest papers by meaning
- BaB-ND: Long-Horizon Motion Planning with Branch-and-Bound and Neural Dynamics (ICLR 2025 · similarity 0.85)
- DeformGen: Dynamics-Based Topology Augmentation for Deformable Manipulation Policy Learning (NeurIPS 2026 · similarity 0.84)
- Stabilizing Reinforcement Learning in Differentiable Multiphysics Simulation (ICLR 2025 · similarity 0.83)
- Implicit Neural Surface Deformation with Explicit Velocity Fields (ICLR 2025 · similarity 0.83)
- Neural Contractive Dynamical Systems (ICLR 2024 · similarity 0.83)
- Contractive Dynamical Imitation Policies for Efficient Out-of-Sample Recovery (ICLR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
