Learning a trajectory using adjoint functions and teacher forcing

Neural Networks

Volumn 5, Issue 3, 1992, Pages 473-484

  Benny Toomarian, Nikzad ^a   Barhen, Jacob ^a,b  

^a CALIFORNIA INSTITUTE OF TECHNOLOGY   (United States)

^b CALIFORNIA INSTITUTE OF TECHNOLOGY   (United States)

Author keywords

Adjoint operators; Recurrent networks; Teacher forcing; Temporal learning; Trajectory learning

Indexed keywords

LEARNING SYSTEMS; MATHEMATICAL TECHNIQUES - ALGORITHMS;

ADJOINT OPERATORS; RECURRENT NETWORKS; TEACHER FORCING; TEMPORAL LEARNING; TRAJECTORY LEARNING;

NEURAL NETWORKS;

EID: 0026685372     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/0893-6080(92)90009-8     Document Type: Article

References (21)

1. Adjoint operator algorithms for faster learning in dynamical neural networks
2. Application of adjoint operators to neural learning
3. Sensitivity theory for nonlinear systems
4. The construction of arbitrary stable dynamics in nonlinear neural networks
5. Convergent activation dynamics in continuous time networks
6. Thermal-hydraulic differential sensitivity theory
7. Identification and control of dynamical systems using neural networks
8. Sensitivity theory for reactor thermal-hydraulic problems
9. Learning state space trajectories in recurrent neural networks
10. Dynamic recurrent neural networks
11. Time dependent adaptive neural networks
12. Learning internal representations by error propagation
13. A real time learning algorithm for recurrent analog neural networks
14. Sensitivity analysis of two-phase flow problems
15. Adjoint operators and nonadiabatic algorithms in neural networks
16. Backpropagation through time: what it does and how to do It
17. A learning algorithm for continually running fully recurrent neural networks
18. A learning algorithm for continually running fully recurrent neural networks
19. Terminal attractors in neural networks