An important issue in Reinforcement Learning (RL) is to accelerate or improve the learning process. In this paper, we study the influence of some RL parameters over the learning speed. Indeed, although RL convergence properties have been widely studied, no precise rules exist to correctly choose the reward function and initial Q-values. Our method helps the choice of these RL parameters within ...

in this paper, the gain in ld-celp speech coding algorithm is predicted using three neural models, that are equipped by genetic and particle swarm optimization (pso) algorithms to optimize the structure and parameters of neural networks. elman, multi-layer perceptron (mlp) and fuzzy artmap are the candidate neural models. the optimized number of nodes in the first and second hidden layers of el...

In this paper, a new hybrid metaheuristic learning algorithm is introduced to choose the best parameters for the multi-criteria decision making algorithm (MCDA) called PROAFTN. PROAFTN requires values of several parameters to be determined prior to classification. These parameters include boundaries of intervals and relative weights for each attribute. The proposed learning algorithm, identifie...

A novel concept learning algorithm named, MICSL: Multiple Iterative Constraint Satisfaction based Learning, is presented. The algorithm utilizes mathematical programming and constraint satisfaction techniques towards uniform representation and management of both data and background knowledge. It offers a flexible enough learning framework and respective services. The representation flexibility ...

The article describes a method combining two widely-used empirical approaches to learning from examples: rule induction and instance-based learning. In our algorithm (RIONA) decision is predicted not on the basis of the whole support set of all rules matching a test case, but the support set restricted to a neighbourhood of a test case. The size of the optimal neighbourhood is automatically ind...

Abstract We propose and develop a new algorithm for trading wind energy in electricity markets, within an online learning optimization framework. In particular, we combine component-wise adaptive variant of the gradient descent with recent advances feature-driven newsvendor model. This results offering approach capable leveraging data-rich environments, while adapting to nonstationary character...

Suppose we are given n coins out of a collection of coins of two distinct weights w0 and w1, true and counterfeit coins, respectively, where d of them are counterfeit coins. Assume we are allowed to weigh subsets of coins in a spring scale. Determine the counterfeit coins in a minimal number of weighing. This problem is equivalent to the following learning problem: Given a linear function f = x...

The rate of convergence of the Generalized Minimum Residual Method (GMRES) applied to the dense linear systems arising from discretization of EFIE integral equation by Method of Moment (MoM) depends heavily by preconditioning. In this work, we evaluate the performances of a simple preconditioner based on the skew hermitian component S of MoM impedance matrix Z for the case of plane wave scatter...

A new, variable structure systems theory based, algorithm has been developed for on-line training of fuzzy-neural networks. Such computationally intelligent structures are widely used for modeling, identification and control of nonlinear dynamic systems. The algorithm is applicable to fuzzy rule-based neural nets of Takagi-Sugeno-Kang type with a scalar output. Its convergence is established an...

This paper proposes an optimized control policy over type one diabetes. Type one diabetes is taken into consideration as a nonlinear model (Augmented Minimal Model), which is implemented in MATLAB-SIMULINK. This Model is developed in consideration of the patient's conditions. There are some uncertainties in the regarded model due to factors such as blood glucose concentration, daily meals or su...