Continuous action reinforcement learning automata and their application to adaptive digital filter design

Filter modeling using gravitational search algorithm

“…Genetic algorithm (GA; Goldberg, 1989), simulated annealing (SA; Kirkpatrick et al, 1983), ant colony optimization (ACO; Dorigo et al, 1996), and particle swarm optimization (PSO; Kennedy and Eberhart, 1995) are four well-known classes of such global optimization methods. The heuristic algorithms are widely used in solving system identification and filter modeling problems (Valarmathi et al, 2009;Chang, 2007;Eftekhari and Katebi, 2008;Chen and Luk, 1999;Howell and Gordon, 2001;Karaboga et al, 2004;Kalinli and Karaboga, 2005;Das and Konar, 2007;Lin et al, 2008).…”

Segmentation with Learning Automata

“…However, the discrete nature of the automata requires the discretization of a continuous parameter space, and the level of quantization tends to reduce the convergence rate. A sequential approach may be adopted (Howell & Gordon, 2000) to overcome such problem by means of an initial coarse quantization. It may be later refined using a re-quantization around the most successful action.…”

“…The continuous action reinforcement learning automata (CARLA) was developed as an extension of the discrete stochastic learning automata for applications involving searching of continuous action space in a random environment (Howell & Gordon, 2000). Several CARLA can be connected in parallel, in a similar manner to discrete automata (Figure 1b environment however, no direct inter-automata communication exist.…”

“…For instance, pattern recognition (Seyed-Hamid, 2008), adaptive control (Zeng et al, 2000) signal processing (Howell & Gordon, 2000) and power systems (Wu, 1995). Recently, some effective algorithms have been proposed for multimodal complex function optimization based on the LA (see (Howell & Gordon, 2000;Thathachar & Sastry, 2002;Zeng & Liu, 2005;Beygi & Meybodi, 2006)). Furthermore, it was shown experimentally that the performance of these optimization algorithms is comparable to or better than the genetic algorithm (GA) in [22].…”

Circle detection on images using learning automata

“…Therefore LA has been employed to solve different sorts of engineering problems, for instance, pattern recognition [17], adaptive control [18], signal processing [19], power systems [20] and computer vision [21]. Other interesting applications for multimodal complex function optimization based on the LA have been proposed in [19,22,23,24], yet showing that their performance is comparable to (GA) [23].…”