Conference Paper

Effects of learning rate on the performance of the population based incremental learning algorithm.

Dept. of Electr. Eng., Univ. of Cape Town, Rondebosch, South Africa
DOI: 10.1109/IJCNN.2009.5179080 Conference: International Joint Conference on Neural Networks, IJCNN 2009, Atlanta, Georgia, USA, 14-19 June 2009
Source: IEEE Xplore


The effect of learning rate (LR) on the performance of a newly introduced evolutionary algorithm called population-based incremental learning (PBIL) is investigated in this paper. PBIL is a technique that combines a simple genetic algorithm (GA) with competitive learning (CL). Although CL is often studied in the context of artificial neural networks (ANNs), it plays a vital role in PBIL in that the idea of creating a prototype vector in learning vector quantization (LVQ) is central to PBIL. In PBIL, the crossover operator of GAs is abstracted away and the role of population is redefined. PBIL maintains a real-valued probability vector (PV) or prototype vector from which solutions are generated. The probability vector controls the random bitstrings generated by PBIL and is used to create other individuals through learning. The setting of the learning rate (LR) can greatly affect the performance of PBIL. However, the effect of the learning rate in PBIL is not yet fully understood. In this paper, PBIL is used to design power system stabilizers (PSSs) for a multi-machine power system. Four cases studies with different learning rate patterns are investigated. These include fixed LR; purely adaptive LR; fixed LR followed by adaptive LR; and adaptive LR followed by fixed LR. It is shown that a smaller learning rate leads to more exploration of the algorithm which introduces more diversity in the population at the cost of slower convergence. On the other hand, a higher learning rate means more exploitation of the algorithm and hence, this could lead to a premature convergence in the case of fixed LR. Therefore, in setting the LR, a trade-off is needed between exploitation and exploration.

Download full-text


Available from: Komla Folly, Oct 05, 2015
1 Follower
56 Reads
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper introduces two improved forms of population based incremental learning (PBIL) algorithm applied to proportional integral derivative (PID) controller and Smith predictor design. Derivative free optimization methods, namely simplex derivative pattern search (SDPS) and implicit filtering (IMF) are used to intensify search mechanism in PBIL algorithm with improved convergence than that of the original PBIL. Although the idea of combining local methods and global methods is not new, this paper focuses application of hybrid heuristics to the vast field of control design especially, control of systems having dead-time. The effectiveness of the controller schemes arrived using the developed algorithms namely simplex derivative pattern search guided population based incremental learning (SDPS-PBIL) and implicit filtering guided population based incremental learning (IMF-PBIL) are demonstrated using unit step set point response for a class of dead-time systems. The results are compared with some existing methods of controller tuning.
    Optimization and Engineering 06/2012; 15(2). DOI:10.1007/s11081-012-9194-2 · 1.23 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: A portfolio selection problem is about finding an optimal scheme to allocate a fixed amount of capital to a set of available assets. The optimal scheme is very helpful for investors in making decisions. However, finding the optimal scheme is difficult and time-consuming especially when the number of assets is large and some actual investment constraints are considered. This paper proposes a new approach based on estimation of distribution algorithms (EDAs) for solving a cardinality constrained portfolio selection (CCPS) problem. The proposed algorithm, termed PBIL-CCPS, hybridizes an EDA called population-based incremental learning (PBIL) algorithm and a continuous PBIL (PBILc) algorithm, to optimize the selection of assets and the allocation of capital respectively. The proposed algorithm adopts an adaptive parameter control strategy and an elitist strategy. The performance of the proposed algorithm is compared with a genetic algorithm and a particle swarm optimization algorithm. The results demonstrate that the proposed algorithm can achieve a satisfactory result for portfolio selection and perform well in searching nondominated portfolios with high expected returns.
    Technologies and Applications of Artificial Intelligence (TAAI), 2010 International Conference on; 12/2010
  • [Show abstract] [Hide abstract]
    ABSTRACT: The aim of the online nonlinear system identification is the accurate modeling of the current local input-output behavior of the plant without using any prior knowledge and offline modeling phase. It is a challenging task for many intelligent systems when used for real-time control applications. In this paper, we propose a novel computationally efficient extended fuzzy functions (EFF) model for system identification of unknown nonlinear discrete-time systems. The main contributions are to introduce an effective quasi-nonlinear model (EFF) and propose adaptive learning rates (ALR) for recursive least squares (RLS) and gradient-descent (GD) methods. The asymptotic convergence of the modeling errors and boundedness of the parameters are proved by using the input-to-state stability (ISS) approach. Numerical simulations are performed for Box–Jenkins gas furnace system and a nonlinear dynamic system. The benefits of its accuracy, stability and simple implementation in practice indicate that EFF model is a promising technique for online identification of nonlinear systems. Copyright © 2010 John Wiley & Sons, Ltd.
    International Journal of Adaptive Control and Signal Processing 02/2011; 25(2):168 - 182. DOI:10.1002/acs.1214 · 1.35 Impact Factor
Show more