Advanced graph-based machine learning reveals cross-sector drivers of decarbonization in the United States and China

Amir Soltani

Applied Energy, 2025, vol. 397, issue C, No S0306261925010980

Abstract: The United States and China, as the world's two largest carbon emitters, play a critical role in global efforts to mitigate climate change. However, there is a notable lack of comprehensive comparative analyses evaluating their decarbonization trajectories across multiple sectors. This study aims to fill this gap by employing advanced machine learning models to analyze and compare how renewable energy adoption, technological advancements, and policy measures have influenced carbon emissions and energy consumption in the United States and China. The nexus of technological innovation and strategic policy implementation is explored to generate actionable insights into the key drivers of power sector decarbonization and the broader clean energy transition. Utilizing a comprehensive dataset covering the power, industry, buildings, and transport sectors, our analysis leverages the strengths of GCN and GAT in capturing complex interdependencies within the data. The findings highlight the pivotal role of innovation and targeted policies in driving significant CO₂ emissions reductions, offering deeper insights into pathways toward net-zero emissions for both countries. This research contributes to the literature by integrating graph-based machine learning approaches to provide a nuanced understanding of feature interactions, which traditional models may overlook, and offers practical recommendations for policymakers and stakeholders engaged in global climate change mitigation efforts. These insights directly inform Article 4 of the Paris Agreement and subsequent Glasgow and Sharm el-Sheikh commitments by quantifying how technology–policy interactions accelerate national emission targets. The graph-based approach also highlights renewable-energy patents and battery breakthroughs as decisive levers, pointing policymakers toward innovation-led decarbonization pathways.

Keywords: Renewable Energy; Decarbonization; Clean Energy transition; CO₂ emissions reduction; Machine learning; Graph convolutional networks; Graph attention networks (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0306261925010980
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:appene:v:397:y:2025:i:c:s0306261925010980

Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/journaldescription.cws_home/405891/bibliographic
http://www.elsevier. ... 405891/bibliographic

DOI: 10.1016/j.apenergy.2025.126368

Access Statistics for this article

Applied Energy is currently edited by J. Yan

More articles in Applied Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().