Research map / Reinforcement learning / Visual and representation RL
Differentiable Model Predictive Control on the GPU
Emre Adabag, Marcus Greiff, John Subosits, Thomas Jonathan Lew
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| replace Riccati recursion with parallel PCG solver (GPU) | DiffMPC vs trajax · RL timing, Problem 1 (GPU) | backward-pass time (s): 1.83 -> 0.32 | Improved |
| DiffMPC vs trajax on nonlinear attitude stabilization | trajax · Nonlinear attitude stabilization | backward time (ms): 505.0 -> 69.3 | Improved |
| run DiffMPC on CPU instead of GPU (vs trajax) | DiffMPC · RL timing, Problem 4 (CPU) | backward-pass time (s): 5.95 -> 10.00 | Hurt |
| warm-start PCG with previous solution (ablation) | DiffMPC · RL timing, PCG tol 1e-12 | forward speedup: 4% | Improved |
| warm-start PCG with looser tolerance 1e-4 (ablation) | DiffMPC · RL timing | forward speedup: 4% -> 11% | Improved |
| use DiffMPC instead of trajax for imitation learning | trajax · Cart-pole imitation learning | training time: approximately 2 times speedup | Improved |
| RL-tune MPC cost and friction parameters with domain randomization | Baseline MPC · Drift with water puddles (simulation) | success rate (%): 70% -> 100% | Improved |
| modify mpc.pytorch to remove serial list comprehension | mpc.pytorch · RL timing, Problem 1 (GPU) | forward time (s): 67.9 -> 1.9 | Improved |
Closest papers by meaning
- Accelerating Visual-Policy Learning through Parallel Differentiable Simulation (NeurIPS 2025 · similarity 0.85)
- Stabilizing Reinforcement Learning in Differentiable Multiphysics Simulation (ICLR 2025 · similarity 0.85)
- Residual-MPPI: Online Policy Customization for Continuous Control (ICLR 2025 · similarity 0.83)
- Differentiable Information Enhanced Model-Based Reinforcement Learning (AAAI 2025 · similarity 0.83)
- Bootstrapped Model Predictive Control (ICLR 2025 · similarity 0.83)
- DOGE-Train: Discrete Optimization on GPU with End-to-End Training (AAAI 2024 · similarity 0.82)
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.
