18 November 2025

A cascading policy learning framework for enhancing power grid resilience

"This paper presents a relevant methodological contribution by proposing a staged reinforcement learning strategy (PPO → TRPO → A2C) aimed at enhancing power-grid resilience. The approach offers a promising direction for stabilizing DRL training in complex control environments. While the results in Grid2Op are encouraging, the lack of broader baselines, scalability analysis, and real-world validation currently limits its generalizability. Nonetheless, the planned release of code and data increases the reproducibility and potential impact of the work."

Carlos Fernandez-Lozano, Section Editor

10 November 2025

FRVC: frame relevance based video compression for surveillance videos using deep learning methods

"The article presents an innovative frame compression mechanism based on the frame importance. Although it is not a revolutionary idea, it is significant enough in its field."

Antonio Jesus Diaz-Honrubia, Handling Editor

7 November 2025

Enhanced Convolutional Neural Network (CNN)-Long Short-Term Memory (LSTM) attention model with adaptive loss for lithium-ion battery state of health estimation

"The paper addresses the current issue of monitoring health status of battery life. Very useful."

Arun Somani, Handling Editor

29 October 2025

Online suicide ideation detection (OnSIDe): a context-aware transfer learning approach using BERT-CNN

"The article presents an interesting data science study for suicide prevention from social media data."

Davide Chicco, Handling Editor

27 October 2025

CIAE: a consistency- and informativeness-aware explanation framework for improving type 2 diabetes prediction and mechanistic insights

"The article presents an interesting application of machine learning to analyze type two diabetes data"

Davide Chicco, Handling Editor

17 October 2025

Pavement defect detection algorithm SSC-YOLO for fusing multiscale spatial channels in YOLOv8

"The SSC-YOLO method, which enhances YOLOv8 by utilizing multiscale spatial channel fusion techniques, is presented in this study as a major breakthrough in pavement fault detection. It is significant because it tackles significant shortcomings in feature preservation and small-object identification, two crucial aspects of real-world infrastructure monitoring. The suggested model's usefulness in autonomous road inspection systems is improved by its consistent gains in mAP and AR across a range of object sizes. Additionally, by providing a lightweight, accurate architecture with enhanced feature extraction and parameter efficiency—balancing deployment practicality with performance in resource-constrained environments—the study advances the discipline."

Doğan Aydın, Handling Editor

16 October 2025

Apple estimation and recognition in complex scenes using YOLO v8

"This work presents a specific and modern analysis of images"

Davide Chicco, Handling Editor

14 October 2025

Design of personalized creation model for cultural and creative products based on evolutionary adaptive network

"The research introduces a novel Evolutionary Adaptive Generative Aesthetic Network (EAGAN), a significant advancement in personalized cultural and creative product design, by integrating text-driven guidance, adaptive style modulation, and evolutionary optimization within the Stable Diffusion architecture"

Muhammad Asif, Handling Editor

10 October 2025

Enhanced piano audio feature recognition: a novel MFCC-based method with F-HRSF and convolutional neural network

"The study significantly refines audio feature extraction by introducing the F-HRSF-based MFCC improvement and leveraging Fisher ratio subband screening, offering a novel framework that enhances recognition accuracy for piano audio. It provides new methodological insights and practical applications, paving the way for further research and innovation in audio analysis and processing technologies"

Muhammad Asif, Handling Editor

2 October 2025

Multiscale attention-based network to enhance detection and classification of autism spectrum disorders using convolutional neural network

"The article proposes an interesting application of deep learning to autism data."

Davide Chicco, Handling Editor