Universal Simulation of Stable Dynamical Systems by Recurrent Neural Nets.
Joshua HansonMaxim RaginskyPublished in: L4DC (2020)
Keyphrases
- dynamical systems
- neural nets
- feed forward
- qualitative simulation
- qualitative differential equations
- dynamic systems
- differential equations
- state space
- phase space
- control theory
- artificial neural networks
- linear systems
- predictive state representations
- nonlinear dynamical systems
- neural network
- back propagation
- learning tasks
- decision making
- feature selection
- partially observable markov decision processes
- artificial intelligence
- agent environment
- image segmentation
- learning algorithm