HiPPO-Prophecy: State-Space Models can Provably Learn Dynamical Systems in Context.
Federico Arangath JosephKilian Konstantin HaefeliNoah LinigerCaglar GulcehrePublished in: CoRR (2024)
Keyphrases
- dynamical systems
- state space
- dynamical models
- dynamical behavior
- differential equations
- dynamic systems
- linear systems
- predictive state representations
- nonlinear dynamical systems
- qualitative simulation
- agent environment
- nonlinear dynamics
- control theory
- discrete dynamical systems
- phase space
- reinforcement learning
- statistical models
- probabilistic model
- ordinary differential equations
- latent variable models
- linear dynamical systems
- heuristic search
- model selection
- particle filter
- dynamic programming