Extending the weightless WiSARD classifier for regression

Filho, Leopoldo Lusquino; Oliveira, Luiz; Filho, Aluizio Lima; Guarisa, Gabriel P.; Felix, Lucca M.; Lima, Priscila M. V.; França, Felipe M. G.

doi:10.1016/j.neucom.2019.12.134

Cited by 8 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…WiSARD (Wilkie, Stonham, Aleksander Recognition Device) weightless neural networks are a type of neural network model that uses a binary pattern and logical rules to perform pattern recognition tasks, rather than numerical weights like traditional neural networks. WiSARD networks store neuron functions using lookup tables, allowing for a simple implementation and fast learning phase [15]. The WiSARD classifier is a fast algorithm that requires simple logical operations that are highly memory efficient since it only stores binary values in random access memory (RAM) cells.…”

Section: Introductionmentioning

confidence: 99%

Weightless neural network-based fault diagnosis in suspension system

Shah,

Venkatesh,

Balaji

et al. 2024

FME Transactions

View full text Add to dashboard Cite

Vehicle suspension systems play a critical role in ensuring passenger comfort and safety. Detecting faults in these systems is vital for maintaining safety, performance, and cost-effectiveness. Traditional inspection methods have limitations, such as visual checks, bounce tests, and alignment assessments. This study explores Wilkie, Stonham, and Aleksander Recognition Device (WiSARD), a weightless neural network (WNN), for suspension fault diagnosis. A WNN model is employed to classify suspension system faults using sensor data. The dataset includes both normal and faulty conditions to train the model. The study assesses WiSARD under various fault conditions, including strut damage, mount failure, worn-out components, and low wheel pressure. Comparative evaluations demonstrate that the approach outperforms other classification techniques, achieving an impressive 95.63% accuracy with a rapid 0.05-second computation time for test data. This WNN-based method proves superior in detecting suspension faults and holds potential as a candidate for real-time vehicle fault diagnosis systems.

show abstract

Section: Introductionmentioning

confidence: 99%

Weightless neural network-based fault diagnosis in suspension system

Shah,

Venkatesh,

Balaji

et al. 2024

FME Transactions

View full text Add to dashboard Cite

show abstract

“…The Wisard is a lightweight online learning supervised learning model based on the N-Tuple classifier [10]. There is an alternative version adapted to perform regression task, called Regression Wisard [11]. It is composed of a set of N RAM nodes.…”

Section: Introductionmentioning

confidence: 99%

Sun Tracking using a Weightless Q-Learning Neural Network

Souza,

Lima,

França

2023

ESANN 2023 Proceesdings

View full text Add to dashboard Cite

Photovoltaic(PV) systems are one of the leading technologies to address climate change. Tracking systems improve energy generation by moving the surface to follow the sun's position however, these methods do not ensure optimal results in cloudy environments. This article proposes a closed-loop control algorithm for tracking based on reinforcement learning and weightless neural networks, compared to an astrological model. The method was applied in a single PV array on a single-axis tracking system, simulated with PVLib. Results showed that the architecture could improve results in cloudy environments but not in a clear-sky situation, as expected for a first approach.

show abstract

“…WiSARD Bagging and WiSARD Boosting proposed here are inspired by the regression ensembles presented in the paper[14] by the same authors of this work and the WiSARD Borda Count is an unpublished strategy originally presented in the PhD thesis of the first author of this work[13].…”

mentioning

confidence: 99%

WiSARD-based Ensemble Learning

Lusquino Filho,

França,

Lima

2023

ESANN 2023 Proceesdings

View full text Add to dashboard Cite

Weightless neural networks are recognized for their online learning capacity and competitive performance with the state-of-the-art in different scenarios. Despite this, the literature has not adequately explored the potential of classification ensembles based on these models and their unique characteristics. This study introduces three types of ensembles based on the WiSARD weightless model and evaluates their effectiveness. The results show that these ensembles significantly improve accuracy compared to the WiSARD model and its ClusWiSARD extension, with a reasonable increase in computational cost. Furthermore, using ensembles eliminates the need for time-consuming tie-break policies of traditional WiSARD models.

show abstract

Extending the weightless WiSARD classifier for regression

Cited by 8 publications

References 14 publications

Weightless neural network-based fault diagnosis in suspension system

Weightless neural network-based fault diagnosis in suspension system

Sun Tracking using a Weightless Q-Learning Neural Network

WiSARD-based Ensemble Learning

Contact Info

Product

Resources

About