A new algorithm for training of nonlinear optimal neuro-controllers (in the form of the model-free, action-dependent, adaptive critic paradigm). Overcomes problems with existing stochastic backpropagation training: need for data storage, parameter shadowing and poor convergence, offering significant benefits for online applications.
|Number of pages||5|
|Journal||IEEE Transactions on Systems, Man and Cybernetics Part B|
|Publication status||Published - Apr 2005|
ASJC Scopus subject areas
- Control and Systems Engineering
- Artificial Intelligence
- Human-Computer Interaction