Stochastic learning-based weak estimation of multinomial random variables and its applications to pattern recognition in non-stationary environments

Oommen, B. John; Rueda, Luis

doi:10.1016/j.patcog.2005.09.007

Cited by 76 publications

(136 citation statements)

References 22 publications

Supporting

Mentioning

131

Contrasting

Order By: Relevance

“…In other words, the SLWE will provide good results even if the distribution parameters change after 50 steps. The experimental results in [5] demonstrated a good performance achieved by using the SLWE in dynamic environments.…”

Section: Stochastic Learning Weak Estimator (Slwe)mentioning

confidence: 72%

“…Similar to the binomial case, the authors of [5] explicitly derived the dependence of E [P (n + 1)] on E [P (n)], demonstrating the ergodic nature of the Markov matrix. The paper 4 also derived two explicit results concerning the convergence of the expected vector P (.)…”

Section: Stochastic Learning Weak Estimator (Slwe)mentioning

confidence: 98%

“…Using the principles of stochastic learning, Oommen and Reuda [5] proposed a strategy to solve the problem of estimating the parameters of a binomial or multinomial distribution efficiently in non-stationary environments. This method is referred to as the SLWE, where the convergence of the estimate is "weak", i.e., with respect to the first and second moments.…”

Section: Stochastic Learning Weak Estimator (Slwe)mentioning

confidence: 99%

“…This feature of "weak" estimators makes these approaches the most effective methods for estimation in non-stationary environments. In this work, we will employ a particular family of weak estimators, referred to as Stochastic Learning Weak Estimation (SLWE) methods [5], for classification in non-stationary environments. The SLWE has been successfully used to solve two-class classification problems by Oommen and Rueda [5] by applying it on non-stationary one-dimensional datasets.…”

Section: Introductionmentioning

confidence: 99%

“…In this work, we will employ a particular family of weak estimators, referred to as Stochastic Learning Weak Estimation (SLWE) methods [5], for classification in non-stationary environments. The SLWE has been successfully used to solve two-class classification problems by Oommen and Rueda [5] by applying it on non-stationary one-dimensional datasets. In this article we will study the performance of the SLWE with more complex classification schemes.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

On the Online Classification of Data Streams Using Weak Estimators

Tavasoli

Oommen

Yazidi

2016

Trends in Applied Knowledge-Based Systems and Data Science

Self Cite

View full text Add to dashboard Cite

Abstract. In this paper, we propose a novel online classifier for complex data streams which are generated from non-stationary stochastic properties. Instead of using a single training model and counters to keep important data statistics, the introduced online classifier scheme provides a real-time self-adjusting learning model. The learning model utilizes the multiplication-based update algorithm of the Stochastic Learning Weak Estimator (SLWE) at each time instant as a new labeled instance arrives. In this way, the data statistics are updated every time a new element is inserted, without requiring that we have to rebuild its model when changes occur in the data distributions. Finally, and most importantly, the model operates with the understanding that the correct classes of previously-classified patterns become available at a later juncture subsequent to some time instances, thus requiring us to update the training set and the training model. The results obtained from rigorous empirical analysis on multinomial distributions, is remarkable. Indeed, it demonstrates the applicability of our method on synthetic datasets, and proves the advantages of the introduced scheme.

show abstract

Section: Stochastic Learning Weak Estimator (Slwe)mentioning

confidence: 72%

Section: Stochastic Learning Weak Estimator (Slwe)mentioning

confidence: 98%

Section: Stochastic Learning Weak Estimator (Slwe)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

On the Online Classification of Data Streams Using Weak Estimators

Tavasoli

Oommen

Yazidi

2016

Trends in Applied Knowledge-Based Systems and Data Science

Self Cite

View full text Add to dashboard Cite

show abstract

On optimizing firewall performance in dynamic networks by invoking a novel swapping window–based paradigm

Mohan

Yazidi

Feng

et al. 2018

Int J Communication

View full text Add to dashboard Cite

Designing and implementing efficient firewall strategies in the age of the internet of things is far from trivial. This is because, as time proceeds, an increasing number of devices will be connected, accessed, and controlled on the internet.Additionally, an ever-increasingly amount of sensitive information will be stored on various networks. A good and efficient firewall strategy will attempt to secure this information and to also manage the large amount of inevitable network traffic that these devices create. The goal of this paper is to propose a framework for designing optimized firewalls for the internet of things. This paper deals with 2 fundamental challenges/problems encountered in such firewalls.The first problem is associated with the so-called rule matching time problem.Here, we propose a simple condition for performing the swapping of the firewall's rules; using which, we can guarantee the firewall's consistency and integrity and also ensure a greedy reduction in the matching time. Unlike the state of the art, our swapping condition considers rules that are not necessarily consecutive, using a novel concept referred to as a "swapping window." The second contribution of our paper is a novel "batch"-based traffic estimator that provides network statistics to the firewall placement optimizer. The traffic estimator is a subtle but modified batch-based embodiment of the Stochastic Learning Weak Estimator. Further, by performing a rigorous suite of experiments, we demonstrate that both algorithms are capable of optimizing the constraints imposed for obtaining an efficient firewall. KEYWORDSbatch update, firewall optimization, learning automata, matching time, nonstationary environments, weak estimators John Oommen is a Chancellor's Professor, Fellow of the IEEE, and Fellow of the IAPR. He is also an Adjunct Professor at Agder University College in Grimstad, Norway. Int J Commun Syst. 2018;31:e3773. wileyonlinelibrary.com/journal/dac Wild card 192.0.* Address range 192.0.0.0-192.0.255.255 Range 192.0.2.2-192.0.2.150 The given range of addresses Single 192.0.2.2 A single given IP address

show abstract

On Allocating Limited Sampling Resources Using a Learning Automata-based Solution to the Fractional Knapsack Problem

Granmo

Oommen

Advances in Soft Computing

View full text Add to dashboard Cite

Stochastic learning-based weak estimation of multinomial random variables and its applications to pattern recognition in non-stationary environments

Cited by 76 publications

References 22 publications

On the Online Classification of Data Streams Using Weak Estimators

On the Online Classification of Data Streams Using Weak Estimators

On optimizing firewall performance in dynamic networks by invoking a novel swapping window–based paradigm

On Allocating Limited Sampling Resources Using a Learning Automata-based Solution to the Fractional Knapsack Problem

Contact Info

Product

Resources

About