Filters








6 Hits in 4.4 sec

Training Reinforcement Neurocontrollers Using the Polytope Algorithm [article]

A. Likas, I. E. Lagaris
1998 arXiv   pre-print
A new training algorithm is presented for delayed reinforcement learning problems that does not assume the existence of a critic model and employs the polytope optimization algorithm to adjust the weights  ...  Experimental results from the application of the method to the pole balancing problem indicate improved training performance compared with critic-based and genetic reinforcement approaches.  ...  The Polytope Algorithm The Polytope algorithm belongs to the class of direct search methods for nonlinear optimization.  ... 
arXiv:cs/9812002v1 fatcat:t74icalwtnd5nfz2l6yskjlkhq

Neural network training and simulation using a multidimensional optimization system

A. Likas, D. A. Karras, I. E. Lagaris
1998 International Journal of Computer Mathematics  
A new approach is presented to neural network simulation and training that is based on the use of general purpose optimization software.  ...  This approach requires that the training problem should be formulated as theminimization of acost functionof the network weights.  ...  A STRATEGY FOR TRAINING REINFORCEMENT NEUROCONTROLLERS Another learning category where the MERLIN optimization system has been proved very useful is the case of delayed reinforcement learning.  ... 
doi:10.1080/00207169808804651 fatcat:7oqgcelzp5gxbmw2cckmjfnysi

Optimization of Interplanetary Solar Sailcraft Trajectories Using Evolutionary Neurocontrol

Bernd Dachwald
2004 Journal of Guidance Control and Dynamics  
., Reinforcement Learning, MIT Press, Cambridge, MA, 1998, pp. 3-23. '3Likas, A., and Lagaris, I., “Training Reinforcement Neurocontrollers Using the Polytope Algorithm,” Neural Processing Letters, Vol  ...  Reinforcement Learning and Neurocontrol Learning algorithms for ANNs that rely on a training set fail when the correct output for a given input is not known.  ... 
doi:10.2514/1.9286 fatcat:umqwziegjvh37nkqx7r6266y3y

Verification for Machine Learning, Autonomy, and Neural Networks Survey [article]

Weiming Xiang and Patrick Musau and Ayana A. Wild and Diego Manzanas Lopez and Nathaniel Hamilton and Xiaodong Yang and Joel Rosenfeld and Taylor T. Johnson
2018 arXiv   pre-print
Recently, the formal methods and formal verification community has developed methods to characterize behaviors in these LECs with eventual goals of formally verifying specifications for LECs, and this  ...  A generalized dynamic backpropagation is used for the training of the parameters of the Diagonal Recurrent NeuroIdentifier (DRNI) and Diagonal Recurrent NeuroController (DRNC).  ...  Figure 1 shows a visualization of the polytopic operations using ReLU functions.  ... 
arXiv:1810.01989v1 fatcat:a5ax66lsxbho3fuxuh55ypnm6m

Page 72 of Journal of Guidance, Control, and Dynamics Vol. 27, Issue 1 [page]

Journal of Guidance, Control, and Dynamics  
., Reinforcement Learning, MIT Press, Cambridge, MA, 1998, pp. 3-23. '3Likas, A., and Lagaris, I., “Training Reinforcement Neurocontrollers Using the Polytope Algorithm,” Neural Processing Letters, Vol  ...  The most commonly used activation function for feed- forward networks is the sigmoid s, : R-+ (0, 1), defined by sy(x) = 1/(1 +e) (Al) where the constant y defines the slope of the function.  ... 

Table of contents

2009 2009 17th Mediterranean Conference on Control and Automation  
The identified neural network (NN) model is validated using the following three different validation algorithms: (1) one-step ahead cross-validation of the training and test data predicted by the trained  ...  The algorithm has been applied to the temperature control of a fluidized bed furnace reactor of the steam deactivation unit of a fluid catalytic cracking (FCC) pilot plant used to evaluate catalyst performance  ...  Its output is used as an error signal by a stable on-line learning algorithm to update the parameters of the neurocontroller.  ... 
doi:10.1109/med.2009.5164498 fatcat:bi37lbkbhfaihj7lbc64vvplo4