Extended Expectation Maximization for Under-Fitted Models

Rekavandi, Aref Miri; Seghouane, Abd‐Krim; Boussaid, Farid; Bennamoun, Mohammed

doi:10.1109/icassp49357.2023.10095526

Cited by 3 publications

(1 citation statement)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The number of radial neurons then needs to be established. This task can be performed, e.g., through the use of clustering algorithms, such as Growing Neural Gas (GNG) [57] or Expectation Maximization (EM) [58] algorithms. The output of each of these neurons is dependent on two parameters: the Euclidean distance between the input variable and its center vector, and the width of the function.…”

Section: Principles Of Operation Of Radial Basis Function Neural Networkmentioning

confidence: 99%

Adaptive Sliding Mode Control Based on a Radial Neural Model Applied for an Electric Drive with an Elastic Shaft

Kaczmarczyk,

Stanislawski,

Szrek

et al. 2024

Energies

View full text Add to dashboard Cite

External disturbances, uncertainties, and nonlinear behavior are problems that are commonly encountered by control system designers. In order to save on energy and materials, mechanical structures have become lighter and more flexible, which only exacerbates the control problem. To resolve this issue, robust and adaptive control strategies have been proposed and have recently gained a lot of interest in modern scientific literature. This article proposes a combination of both approaches: a sliding mode—radial basis function neural network controller applied to an electrical drive with a sophisticated mechanical structure. The proposed sliding surface provides robustness against parameter uncertainties, while the neural network adjusts itself to the current state of the drive and mitigates the oscillations resulting from the elastic connection with the load machine. This article proves the stability of the proposed control algorithm in the sense of Lyapunov, provides an in-depth numerical analysis, and compares those results with the experimental tests. The algorithm was implemented in a 1103 dSPACE fast-prototyping card and was used to control a 0.5 kW DC motor connected to the load machine by a long (thin) steel shaft.

show abstract

Section: Principles Of Operation Of Radial Basis Function Neural Networkmentioning

confidence: 99%

Adaptive Sliding Mode Control Based on a Radial Neural Model Applied for an Electric Drive with an Elastic Shaft

Kaczmarczyk,

Stanislawski,

Szrek

et al. 2024

Energies

View full text Add to dashboard Cite

show abstract

Analysis and evaluation of explainable artificial intelligence on suicide risk assessment

Tang,

Miri Rekavandi,

Rooprai

et al. 2024

Sci Rep

View full text Add to dashboard Cite

This study explores the effectiveness of Explainable Artificial Intelligence (XAI) for predicting suicide risk from medical tabular data. Given the common challenge of limited datasets in health-related Machine Learning (ML) applications, we use data augmentation in tandem with ML to enhance the identification of individuals at high risk of suicide. We use SHapley Additive exPlanations (SHAP) for XAI and traditional correlation analysis to rank feature importance, pinpointing primary factors influencing suicide risk and preventive measures. Experimental results show the Random Forest (RF) model is excelling in accuracy, F1 score, and AUC (>97% across metrics). According to SHAP, anger issues, depression, and social isolation emerge as top predictors of suicide risk, while individuals with high incomes, esteemed professions, and higher education present the lowest risk. Our findings underscore the effectiveness of ML and XAI in suicide risk assessment, offering valuable insights for psychiatrists and facilitating informed clinical decisions.

show abstract