Monocular Object-Level SLAM Enhanced by Joint Semantic Segmentation and Depth Estimation by Ruicheng Gao, Yue Qi

“…The multi-task network, called JSDNet, is designed to predict depth and semantic segmentation simultaneously, with four contributions that include depth discretization, feature fusion, a weight-learned loss function, and semantic consistency optimization. …”
Get full text

A New Hydrothermal Coupling Model for Predicting Temperature Profiles: Impact of Porosity, Density and Meteorological Factors by D. Bennaceur, N. Laredj, M. Maliki, H. Missoum

“…Accurately predicting soil temperature variation plays a crucial role in the design and optimization of geothermal applications, as it directly impacts the efficiency and performance of geothermal energy systems. …”
Get full text

Presenting a Stochastic Model of Simultaneous Planning Problem of Distribution and Subtransmission Network Development considering the Reliability and Security Indicators by Seyed Hossein Moayed, Hosein Hasan Shahi, Mohammadreza Akbarizadeh, Morteza Jadidoleslam, Arash Aghatehrani, Afshin Pirouzi

“…The aforementioned problem is in the form of an optimization problem whose objective function is to minimize the costs of construction, replacement, operation, maintenance, reliability, and security. …”
Get full text

A Learned Reduced-Rank Sharpening Method for Multiresolution Satellite Imagery by Sveinn E. Armannsson, Magnus O. Ulfarsson, Jakob Sigurdsson

“…Our method combines traditional model-based methods with neural network optimization techniques. Our method solves the same optimization problem as traditional model-based methods while leveraging neural network optimization techniques through a customized U-Net architecture and specialized loss function. …”
Get full text

ELFA-Log: Cross-System Log Anomaly Detection via Enhanced Pseudo-Labeling and Feature Alignment by Xiaowei Zhao, Kaiwei Guo, Mingting Huang, Shaojian Qiu, Lu Lu

“…Additionally, a distance-based loss function optimizes the shared representation of cross-system log features. …”
Get full text

Frequency Enhanced Dynamic Graph Convolutional Networks for Traffic Flow Forecasting by Liyuan Wang, Jiafeng Zhuang, Shuo Ma, Hai Lin

“…Additionally, the proposed method integrates the Fourier transform into the loss function to mitigate label autocorrelation and improve long-term forecasting accuracy. …”
Get full text

Research on the Method of Crop Pest and Disease Recognition Based on the Improved YOLOv7-U-Net Combined Network by Wenchao Xiang, Zitao Du, Xinran Liu, Zehui Lu, Yuna Yin

“…Experimental results show the improved algorithm achieves 97.49% detection accuracy, with mean average precision (mAP) reaching 96.91% and detection speed increasing to 90.41 FPS. The loss function of the improved U-Net network decreases towards 0 with training iterations, validating its effectiveness. …”
Get full text

Identification of genetically engineered strategies to manipulate nano-platforms presenting immunotherapeutic ligands for alleviating primary ovarian insufficiency progression by Guannan Zhou, Yuanyuan Gu, Menglei Zhang, Jingxin Ding, Guanming Lu, Keqin Hua, Fang Shen

“…Abstract Primary ovarian insufficiency (POI) is a pathological condition characterized by the early loss of functional ovarian follicles, leading to infertility and systemic consequences affecting reproductive, skeletal, cardiovascular, and neurocognitive helath. …”
Get full text

Generalized Score Comparison-Based Learning Objective for Deep Speaker Embedding by Min Hyun Han, Sung Hwan Mun, Nam Soo Kim

“…Furthermore, we propose a generalized loss function for score comparison-based learning, encompassing many conventional training losses and regularization techniques. …”
Get full text

RDW-YOLO: A Deep Learning Framework for Scalable Agricultural Pest Monitoring and Control by Jiaxin Song, Ke Cheng, Fei Chen, Xuecheng Hua

“…Third, an enhanced Wise-Wasserstein IoU (WWIoU) loss function optimizes the matching mechanism and improves bounding-box regression. …”
Get full text

Multi-WSN Simulator with Log Data for Efficient Sensing on Internet of Things by Young-Sik Jeong, Hyun-Woo Kim, Neil Y. Yen, Jong Hyuk Park

“…Moreover, even though sensor nodes have the same function, there are different kinds; therefore, it is difficult to select and place optimal sensor nodes. …”
Get full text

Enhancing Early Exit Performance With Uncertainty-Aware Training in Convolutional Neural Networks for Image Classification by Thanu Kurian, Somasundaram Thangam

“…To overcome the downfall, we propose uncertainty estimation in the loss function as an additional parameter for classification. …”
Get full text

Opportunities and Challenges for E-health in Hip and Knee Arthroplasty - A Summary of Recent Literature Review by José Seguro, Márcia Santos

“…<strong>Conclusions:</strong> E-health has promising potential for optimizing arthroplasty care.…”
Get full text

Pre Hoc and Co Hoc Explainability: Frameworks for Integrating Interpretability into Machine Learning Training for Enhanced Transparency and Performance by Cagla Acun, Olfa Nasraoui

“…Pre hoc uses the white-box model to regularize the black-box model, while co hoc jointly optimizes both models with a shared loss function. …”
Get full text

Three-Dimensional RF Source Localization Using Reflection and an Improved Particle Filter by Saeid Haidari, Alireza Hoseinpour

“…In the first step of the localization, an unmanned aerial vehicle (UAV) is used to obtain AOA and locate the three-dimensional reflection using a two-dimensional map. Then, the path loss function is used and the reflection angle alongside the distance between the receiver and RF source is estimated based on RSSI. …”
Get full text

PLANNING THE GROWTH OF THE REGIONAL ECONOMY BASED ON SIMULATION OF INVESTMENT STRATEGIES by I. S. Mezhov, E. V. Kletskova

“…The above approach is justified because, on the one hand, there are not enough publications on the theoretical principles of managing an inter-sectoral economic complex at the regional level in Russia, on the other, the majority of regions from year to year have been suffering from a deficit of investments and regional budgets, loss of production and jobs. A model is proposed that can be formed on the basis of a multiparametric representation of the industrial and technological function of the industry without being reduced only to cost factors as in the inter-sectoral balance sheets. …”
Get full text

The analysis of entrepreneurship evaluation system for talent cultivation in artistic creativity and animation under artificial intelligence by Hao Xie, Zijian Zhao

“…The performance of the model is evaluated, revealing that it outperforms existing technologies in indicators such as loss function value, fitting effect, accuracy, precision, recall, and F1 score, with an accuracy of up to 96.30%. …”
Get full text

A novel two-stage feature selection method based on random forest and improved genetic algorithm for enhancing classification in machine learning by Junyao Ding, Jianchao Du, Hejie Wang, Song Xiao

“…Then, the improved genetic algorithm is used to search for the global optimal feature subset further. This process introduces a multi-objective fitness function to guide the feature subset, minimizing the number of features in the subset while enhancing classification accuracy. …”
Get full text

Neural radiance fields assisted by image features for UAV scene reconstruction by Zhihong Chen, Xueyun Chen, Chenghong Ye, Shaojie Wu, Xiang Wu

“…To further improve synthesis quality, we also refine the loss function used during training to better guide network optimization. …”
Get full text

Road-Adaptive Precise Path Tracking Based on Reinforcement Learning Method by Bingheng Han, Jinhong Sun

“…Next, based on the generated reference path, the current state of the vehicle, and the vehicle motor energy efficiency diagram, the optimal speed is calculated in real time, and the vehicle dynamics preview point at the future moment—specifically, the look-ahead distance—is predicted. …”
Get full text