Bilinear Mamba-Koopman Neural MPC introduces control-dependent coupling in latent dynamics as a strict generalization of standard Koopman MPC, improving forecasting and closed-loop robustness on time-varying CartPole and RSCP tasks with minimal added parameters.
Proceedings of the American Mathematical Society , volume =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Bilinear Mamba-Koopman Neural MPC for Varying Dynamics
Bilinear Mamba-Koopman Neural MPC introduces control-dependent coupling in latent dynamics as a strict generalization of standard Koopman MPC, improving forecasting and closed-loop robustness on time-varying CartPole and RSCP tasks with minimal added parameters.