New KRPRVFL model enhances neural network robustness against noisy data

Researchers have developed a new neural network model called KRPRVFL, designed to improve the robustness of Random Vector Functional Link (RVFL) networks. This new model addresses the sensitivity of traditional RVFL networks to noisy data, outliers, and imbalanced datasets by incorporating a kernel risk-sensitive mean p-power (KRP) criterion. The KRPRVFL model adaptively reduces the impact of unreliable samples during training and includes a collaborative learning mechanism for enhanced stability in complex environments. Experiments on benchmark datasets indicate that KRPRVFL offers superior accuracy and robustness compared to existing models. AI

IMPACT Introduces a more robust neural network architecture for classification tasks, potentially improving performance in real-world applications with noisy data.

RANK_REASON Academic paper detailing a new model architecture and its experimental validation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• IArxiv
• Influence Flower
• KEEL
• Kernel Risk-Sensitive Mean p-Power
• KRPRVFL
• ScienceCast
• University of California, Irvine

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →