Introduction: New digital infrastructure (NDI) is a key driver for agricultural green development (AGD). However, its underlying mechanisms and heterogeneous impacts remain to be explored. This study focuses on agricultural green total factor productivity (AGTFP) in China, aiming to elucidate the direct effects of NDI on AGD, its mediating pathways, and geographical heterogeneity, thereby providing theoretical evidence and policy implications for the green transformation of agriculture.Methods: Based on panel data from 30 provinces in China from 2006 to 2022, this study employs the global Malmquist-Luenberger (GML) index and the slack-based measure (SBM) model to measure AGTFP and constructs a comprehensive NDI indicator using principal component analysis. The study systematically examines the direct impact of NDI on AGD, the mediating effects of technological and land factors, and the moderating role of labor quality through fixed-effects models, instrumental variable regression, and moderated-mediation effect models. Heterogeneity analysis is also conducted through subgroup regression to explore geographical and institutional differences.Results: NDI significantly enhances AGTFP, a conclusion that is robust to endogeneity and sensitivity tests. technology level and farmland quality are the dual mediating pathways through which NDI drives AGD. Labor quality amplifies the green empowerment effect of NDI by strengthening the synergistic effects of technological and land factors. Heterogeneity analysis shows that the effect of NDI is more pronounced in ecologically fragile areas in the northwest, as well as in regions with low marketization and strict environmental regulations.Discussion: NDI promotes the green transformation of agriculture by optimizing resource utilization efficiency and driving technological innovation. However, it requires matching with improved labor quality and region-specific policies. It is recommended to prioritize the deployment of intelligent monitoring facilities in ecologically fragile areas to compensate for institutional shortcomings through digital technology; and to promote precision agronomic systems in intensive agricultural areas to reduce resource dependence and accelerate the diffusion of green technologies through market mechanisms. The limitations of this study include the omission of spatial spillover effects and potential biases in the calculation of agricultural value-added. Future research could further explore these aspects by incorporating spatial econometric models.

The new digital infrastructure has become an important basis for the green development of agriculture. This paper selects provincial sample data from 2006 to 2022, constructs an indicator system to measure new digital infrastructure and agricultural green total factor productivity, and explores how new digital infrastructure can promote agricultural green development. The results show that: (1) the new digital infrastructure has a significant role in promoting the green development of agriculture, and this conclusion is still valid after processing endogeneity and heteroscedasticity; (2) Technology and land factors play an intermediary role in the promotion of green agricultural development by the new digital infrastructure, that is, the new digital infrastructure helps to improve the quality and technical level of cultivated land and thus promote green agricultural development; (3) Labor force factors optimize the mediating role of technology factors, that is, the higher the quality of labor force, the stronger the role of the new digital infrastructure in promoting the green development of agriculture through the technological path; (4) The effect of new digital infrastructure on the green development of agriculture is obviously heterogeneous, and the effect is more obvious in northwest China and the regions with lower marketization level, higher environmental regulation level. This study is helpful to clarify the relationship between new digital infrastructure and agricultural green development in China, and provide corresponding policy suggestions for promoting regional agricultural green development.

This paper proposes an adaptive risk control deep Q network (ARC-DQN) model based on deep reinforcement learning, focusing on the optimization of futures and foreign exchange trading strategies. The model introduces risk compensation factors and adaptive weight adjustment mechanisms to incorporate the dynamic balance between returns and risks into the decision-making process, achieving robust trading in a highly volatile market environment based on the standard DQN framework. The model represents market data with a multi-dimensional state vector, including prices, technical indicators, and risk indicators; the reward function is adjusted through a risk compensation mechanism during the action selection process, thereby effectively controlling volatility and maximum drawdown. The experiment uses historical data from the futures and foreign exchange markets for simulation. The results show that the ARC-DQN model improves the cumulative rate of return by about 22% compared with the traditional DQN, the Sharpe ratio increases by about 18%, the maximum drawdown decreases by about 15%, and the system response time is less than 2 s.