Research map / Learning theory and algorithms / Prediction and decision losses
Meta-Adaptive Nonlinear Control: Theory and Algorithms
Guanya Shi, Kamyar Azizzadenesheli, Michael O'Connell, Soon-Jo Chung, Yisong Yue
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 |
|---|---|---|---|
| Use OMAC instead of baseline adaptive control | Adaptive control · Inverted pendulum | ACE: 0.311\pm0.112 -> 0.093\pm0.027 | Improved |
| Use OMAC instead of baseline adaptive control | Adaptive control · 6-DoF quadrotor | ACE: 0.283\pm0.043 -> 0.141\pm0.024 | Improved |
| Use the deep representation model instead of the convex superposition model (ablation) | OMAC · Inverted pendulum | ACE: 0.147\pm0.047 -> 0.093\pm0.027 | Improved |
| Use the deep representation model instead of the convex superposition model (ablation) | OMAC · 6-DoF quadrotor | ACE: 0.251\pm0.043 -> 0.141\pm0.024 | Improved |
| Use the bi-convex model instead of the convex superposition model (ablation) | OMAC · Inverted pendulum | ACE: 0.147\pm0.047 -> 0.129\pm0.044 | Improved |
| Use the bi-convex model instead of the convex superposition model (ablation) | OMAC · 6-DoF quadrotor | ACE: 0.251\pm0.043 -> 0.150\pm0.019 | Improved |
| Use the deep representation model instead of the bi-convex model (ablation) | OMAC · 6-DoF quadrotor | ACE: 0.150\pm0.019 -> 0.141\pm0.024 | No clear effect |
| Learn and update a shared representation across environments | OMAC versus baseline adaptive control · Inverted pendulum and 6-DoF quadrotor | ACE | Improved |
| Use a convex superposition model for aerodynamic force prediction (ablation) | OMAC · 6-DoF quadrotor | ACE: 0.283\pm0.043 -> 0.251\pm0.043 | No clear effect |
Closest papers by meaning
- Decoupling Meta-Reinforcement Learning with Gaussian Task Contexts and Skills (AAAI 2024 · similarity 0.85)
- Regret Analysis of Multi-task Representation Learning for Linear-Quadratic Adaptive Control (AAAI 2025 · similarity 0.84)
- World Models as Reference Trajectories for Rapid Motor Adaptation (NeurIPS 2025 · similarity 0.84)
- Adaptive Horizon Actor-Critic for Policy Learning in Contact-Rich Differentiable Simulation (ICML 2024 · similarity 0.84)
- Towards an Information Theoretic Framework of Context-Based Offline Meta-Reinforcement Learning (NeurIPS 2024 · similarity 0.84)
- Meta-Controller: Few-Shot Imitation of Unseen Embodiments and Tasks in Continuous Control (NeurIPS 2024 · 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.
