“Boosted Multi-Task Learning for Inter-District Collaborative Load Forecasting”
Haizhou Liu, Xuan Zhang, Hongbin Sun, M. Shahidehpour
IEEE Transactions on Smart Grid (2024)
This paper introduces a novel framework for collaborative load forecasting across different districts, focusing on Zhuhai city as a case study. The proposed framework operates in two distinct stages: initially, districts collaborate under a federated learning scheme to capture global load patterns, followed by local training to refine district-specific predictions. The framework leverages a probabilistic Gradient-Boosted Regression Tree (GBRT) as the core machine learning algorithm, enabling effective multi-task learning. The study also explores two district withdrawal mechanisms—simultaneous withdrawal for accuracy and dynamic withdrawal for efficiency and incentivization. The results demonstrate the superiority of this collaborative approach, showing significant improvements in forecasting performance across all districts involved.
“Regional Integrity and Corporate Green-Technology Innovation: Evidence from Deadbeat Borrowers in China”
Qihang Xue, Caiquan Bai, Jinmeng Shi, Dequn Cui
Emerging Markets Finance and Trade (2024)
This study investigates the role of regional integrity—an informal institutional factor—in influencing corporate green-technology innovation (GTI) in China. Utilizing data from individuals who defaulted on court orders and green patent-licensing information from listed companies, the researchers find that lower levels of regional integrity significantly impede GTI. The study employs an instrumental-variable approach to address potential endogeneity and confirms the robustness of its findings. The analysis reveals that lower regional integrity reduces firms’ willingness to disclose environmental information and weakens the impact of environmental regulations. This effect is more pronounced in firms where directors and general managers do not hold concurrent posts, in state-owned enterprises, and for utility patents. The paper highlights the importance of enhancing regional integrity to foster corporate innovation in green technologies.
“Influence and Mechanism of Traditional Chinese Medicine Intervention on Cognitive Dysfunction in Patients with Schizophrenia”
Ningbo Yang, Hongxia Hu, Jie Li, Shaoli Shi, Yuling Wei, Yanhong Li, Wenwen Sun
Journal of Biobased Materials and Bioenergy (2024)
This research examines the efficacy of traditional Chinese medicine (TCM) in treating cognitive dysfunction among schizophrenia patients. The study involved both animal experiments and clinical trials. In animal studies, rats treated with Albizia flower flavonoids (AFFG) showed significant improvements in cognition and hippocampal pathology compared to controls. In clinical trials, patients treated with a combination of aripiprazole and a self-formulated TCM decoction exhibited lower PANSS scores and improved quality of life compared to those receiving only aripiprazole. The study suggests that AFFG modulates protein expression, enhances cognitive function, and reduces pathological symptoms, making it a promising adjunctive treatment for schizophrenia. The findings support the potential clinical application of TCM in improving cognitive outcomes in patients with schizophrenia.
No Comments