Generalized pursuit learning schemes: new families of continuous and discretized learning automata

Agache, M.; Oommen, B. John

doi:10.1109/tsmcb.2002.1049608

Cited by 152 publications

(166 citation statements)

References 44 publications

Supporting

Mentioning

166

Contrasting

Order By: Relevance

“…1 This sections presents approaches for to predicting p(t i , n). The first approach is a straightforward solution introduced for comparison purposes.…”

Section: Problem Setting and Modellingmentioning

confidence: 99%

“…Learning Automata (LA) have been used in systems that have incomplete knowledge of the environment in which they operate [1,18,25,26,27,32,39]. The learning mechanism attempts to learn from a stochastic Teacher which models the Environment.…”

Section: Meta-learning Algorithm Based On Learning Automatamentioning

confidence: 99%

“…The goal of LA is to "determine the optimal action out of a set of allowable actions" [1]. The distinguishing characteristic of automata-based learning is that the search for the optimizing parameter vector is conducted in the space of probability distributions defined over the parameter space, rather than in the parameter space itself [38].…”

Section: Meta-learning Algorithm Based On Learning Automatamentioning

confidence: 99%

“…A simple example is the aggregate behavior of many consumers turning on their home heaters simultaneously, within a short time span, as a consequence of a cold weather wave. This aggregate behavior is natural since a drop in temperature affects a large population which might 1 A preliminary version of this paper that does not include the meta-learning algorithm (as well as the theory of learning automata) introduced in this journal paper was presented at AIAI'14, the 2014 IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations, Rhodes, Greece, in September 2014. We are very grateful to the anonymous Referees of the previous version of the paper, whose suggestions greatly improved the quality of this version.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Pattern Recognition Approach for Peak Prediction of Electrical Consumption

Goodwin

Yazidi

2014

Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications

View full text Add to dashboard Cite

Abstract.Demand peaks in electrical power consumptions pose serious challenges for energy companies as these are typically unforeseen and require the net to support abnormally high consumption levels. Such peaks can be regulated in smart energy grids with the introduction of relatively simple techniques such as load balancing and smart pricing strategies. This is, however, difficult in practice because it requires prediction of peaks prior to their actual occurrence.While most studies formulate the problem as an estimation problem, we take a radically different approach and formulate it as a classical pattern recognition problem. Further, the paper applies classification methods to solve the problem and applies these with real-life data from a Norwegian smart grid pilot project. Some of the key findings are that the algorithms can accurately detect 80% of energy consumption peaks up to one week ahead of time. Furthermore, we introduce a novel Learning Automata based approaches for selecting the optimal prediction model from a pool of models in an online fashion.

show abstract

“…1 This sections presents approaches for to predicting p(t i , n). The first approach is a straightforward solution introduced for comparison purposes.…”

Section: Problem Setting and Modellingmentioning

confidence: 99%

Section: Meta-learning Algorithm Based On Learning Automatamentioning

confidence: 99%

Section: Meta-learning Algorithm Based On Learning Automatamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Pattern Recognition Approach for Peak Prediction of Electrical Consumption

Goodwin

Yazidi

2014

Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications

View full text Add to dashboard Cite

show abstract

“…As opposed to this, our scheme resorts to discretizing the probability space [1,5,9,16], and performing a controlled random walk on this discretized space. It is well known in the field of LA that discretized schemes achieve faster convergence speed than continuous schemes [1,8]. By virtue of discretization, our estimator realizes fast adjustments of the running estimates by jumps, and it is thus able to robustly track changes in the parameters of the distribution after a switch has occurred in the environment.…”

Section: State-of-the-artmentioning

confidence: 99%

A Stochastic Search on the Line-Based Solution to Discretized Estimation

Yazidi

Granmo

Oommen

2012

Advanced Research in Applied Artificial Intelligence

View full text Add to dashboard Cite

Abstract. Recently, Oommen and Rueda [11] presented a strategy by which the parameters of a binomial/multinomial distribution can be estimated when the underlying distribution is nonstationary. The method has been referred to as the Stochastic Learning Weak Estimator (SLWE), and is based on the principles of continuous stochastic Learning Automata (LA). In this paper, we consider a new family of stochastic discretized weak estimators pertinent to tracking time-varying binomial distributions. As opposed to the SLWE, our proposed estimator is discretized, i.e., the estimate can assume only a finite number of values. It is well known in the field of LA that discretized schemes achieve faster convergence speed than their corresponding continuous counterparts. By virtue of discretization, our estimator realizes extremely fast adjustments of the running estimates by jumps, and it is thus able to robustly, and very quickly, track changes in the parameters of the distribution after a switch has occurred in the environment. The design principle of our strategy is based on a solution, pioneered by Oommen [7], for the Stochastic Search on the Line (SSL) problem. The SSL solution proposed in [7], assumes the existence of an Oracle which informs the LA whether to go "right" or "left". In our application domain, in order to achieve efficient estimation, we have to first infer (or rather simulate) such an Oracle. In order to overcome this difficulty, we rather intelligently construct an "Artificial Oracle" that suggests whether we are to increase the current estimate or to decrease it. The paper briefly reports conclusive experimental results that demonstrate the ability of the proposed estimator to cope with non-stationary environments with a high adaptation rate, and with an accuracy that depends on its resolution. The results which we present are, to the best of our knowledge, the first reported results that resolve the problem of discretized weak estimation using a SSL-based solution.

show abstract

Fast Learning Automata

2023

Learning Automata and Their Applications to Intelligent Systems

View full text Add to dashboard Cite

Generalized pursuit learning schemes: new families of continuous and discretized learning automata

Cited by 152 publications

References 44 publications

A Pattern Recognition Approach for Peak Prediction of Electrical Consumption

A Pattern Recognition Approach for Peak Prediction of Electrical Consumption

A Stochastic Search on the Line-Based Solution to Discretized Estimation

Fast Learning Automata

Contact Info

Product

Resources

About