Generative AI,represented by GPT(Generative Pre-trained Transformer),is now leading the technological revolution and is reconstructing the journalism and communication industries'ecosystems because of its powerful...Generative AI,represented by GPT(Generative Pre-trained Transformer),is now leading the technological revolution and is reconstructing the journalism and communication industries'ecosystems because of its powerful generative capacity and diverse range of outputs.While GPT is busy revolutionizing and innovating the production of news content,working patterns,and operation modes,it has also given rise to ethical concerns in regard to news authenticity,data security,humanistic values,and other related aspects.Therefore,it is imperative to initiate strategies and approaches,such as establishing a mechanism for verifying information authenticity,enhancing data security and privacy regulations,and instituting an ethical supervision and governance framework for AI,in order to facilitate the systematic advancement of AI-based news production while reinstating public trust.展开更多
The tropospheric delay has a significant impact on high-accuracy positioning of the Global Navigation Satellite System(GNSS).Traditional solutions have their weaknesses.First,the estimation of tropospheric delay as a ...The tropospheric delay has a significant impact on high-accuracy positioning of the Global Navigation Satellite System(GNSS).Traditional solutions have their weaknesses.First,the estimation of tropospheric delay as a state parameter slows the positioning filter's convergence,especially critical for Precise Point Positioning(PPP).Second,correction-based approaches,including empirical model,meteorological model and GNSS network observations,have their corresponding limitations.The empirical model comprises yearly data-based statistics,which ignores high temporal-variation components,leading to decreased correction accuracy.The meteorological model requires real-time local weather observations.One can enable the network method of the expensive regional infrastructure of GNSS stations,of which performance depends on the rover-network geometry.In this study,we enable a real-time tropospheric regional correction service by polynomial coefficients from the Kalman filtering of multisource data,including the Global Pressure and Temperature 2 wet(GPT2w)model,weather observations from the National Oceanic and Atmospheric Administration(NOAA),and GNSS network observations.After discussing the weighting strategy examined by the regional dataset from Zhejiang Province,we evaluate the performance of the proposed fusion approach with post-processed PPP results as references.We obtained the optimal weightings for the corresponding dataset,and the average accuracy for Zenith Tropospheric Delay(ZTD)is 0.43,and 1.20 cm under static,active,and overall weather conditions,respectively.Compared with the real-time GNSS network ZTD solution,our proposed fusion solution is improved by 48.21%,55.20%,and 41.70%,respectively.In conclusion,the proposed approach makes the best of three traditional correction-based methods to provide optimized real-time tropospheric service.展开更多
The pursuit of optimal neural network architectures is foundational to the progression of Neural Architecture Search (NAS). However, the existing NAS methods suffer from the following problem using traditional search ...The pursuit of optimal neural network architectures is foundational to the progression of Neural Architecture Search (NAS). However, the existing NAS methods suffer from the following problem using traditional search strategies, i.e., when facing a large and complex search space, it is difficult to mine more effective architectures within a reasonable time, resulting in inferior search results. This research introduces the Generative Pre-trained Transformer NAS (GPT-NAS), an innovative approach designed to overcome the limitations which are inherent in traditional NAS strategies. This approach improves search efficiency and obtains better architectures by integrating GPT model into the search process. Specifically, we design a reconstruction strategy that utilizes the trained GPT to reorganize the architectures obtained from the search. In addition, to equip the GPT model with the design capabilities of neural architecture, we propose the use of the GPT model for training on a neural architecture dataset. For each architecture, the structural information of its previous layers is utilized to predict the next layer of structure, iteratively traversing the entire architecture. In this way, the GPT model can efficiently learn the key features required for neural architectures. Extensive experimental validation shows that our GPT-NAS approach beats both manually constructed neural architectures and automatically generated architectures by NAS. In addition, we validate the superiority of introducing the GPT model in several ways, and find that the accuracy of the neural architecture on the image dataset obtained from the search after introducing the GPT model is improved by up to about 9%.展开更多
To improve the applicability of the global pressure and temperature 2 wet(GPT2w)model in estimating the weighted mean temperature in China and adjacent areas,the error compensation technology based on the neural netwo...To improve the applicability of the global pressure and temperature 2 wet(GPT2w)model in estimating the weighted mean temperature in China and adjacent areas,the error compensation technology based on the neural network was proposed,and a total of 374800 meteorological profiles measured from 2006 to 2015 of 100 radiosonde stations distributed in China and adjacent areas were used to establish an enhanced empirical model for estimating the weighted mean temperature in this region.The data from 2016 to 2018 of the remaining 92 stations in this region was used to test the performance of the proposed model.Results show that the proposed model is about 14.9%better than the GPT2w model and about 7.6%better than the Bevis model with measured surface temperature in accuracy.The performance of the proposed model is significantly improved compared with the GPT2w model not only at different height ranges,but also in different months throughout the year.Moreover,the accuracy of the weighted mean temperature estimation is greatly improved in the northwestern region of China where the radiosonde stations are very rarely distributed.The proposed model shows a great application potential in the nationwide real-time ground-based global navigation satellite system(GNSS)water vapor remote sensing.展开更多
基金the phased achievement of the subject“Research on Deepening the Construction of‘Smart Chengdu’and Enhancing the Governance Efficiency of Megacities” (2023CS120)a key project under Chengdu’s Philosophy and Social Science Planning Program for 2023。
文摘Generative AI,represented by GPT(Generative Pre-trained Transformer),is now leading the technological revolution and is reconstructing the journalism and communication industries'ecosystems because of its powerful generative capacity and diverse range of outputs.While GPT is busy revolutionizing and innovating the production of news content,working patterns,and operation modes,it has also given rise to ethical concerns in regard to news authenticity,data security,humanistic values,and other related aspects.Therefore,it is imperative to initiate strategies and approaches,such as establishing a mechanism for verifying information authenticity,enhancing data security and privacy regulations,and instituting an ethical supervision and governance framework for AI,in order to facilitate the systematic advancement of AI-based news production while reinstating public trust.
基金supported by the National Natural Science Foundation of China under[Grants 42004019 and 41874033].
文摘The tropospheric delay has a significant impact on high-accuracy positioning of the Global Navigation Satellite System(GNSS).Traditional solutions have their weaknesses.First,the estimation of tropospheric delay as a state parameter slows the positioning filter's convergence,especially critical for Precise Point Positioning(PPP).Second,correction-based approaches,including empirical model,meteorological model and GNSS network observations,have their corresponding limitations.The empirical model comprises yearly data-based statistics,which ignores high temporal-variation components,leading to decreased correction accuracy.The meteorological model requires real-time local weather observations.One can enable the network method of the expensive regional infrastructure of GNSS stations,of which performance depends on the rover-network geometry.In this study,we enable a real-time tropospheric regional correction service by polynomial coefficients from the Kalman filtering of multisource data,including the Global Pressure and Temperature 2 wet(GPT2w)model,weather observations from the National Oceanic and Atmospheric Administration(NOAA),and GNSS network observations.After discussing the weighting strategy examined by the regional dataset from Zhejiang Province,we evaluate the performance of the proposed fusion approach with post-processed PPP results as references.We obtained the optimal weightings for the corresponding dataset,and the average accuracy for Zenith Tropospheric Delay(ZTD)is 0.43,and 1.20 cm under static,active,and overall weather conditions,respectively.Compared with the real-time GNSS network ZTD solution,our proposed fusion solution is improved by 48.21%,55.20%,and 41.70%,respectively.In conclusion,the proposed approach makes the best of three traditional correction-based methods to provide optimized real-time tropospheric service.
基金supported by the National Nature Science Foundation of China(No.62106161)the Fundamental Research Funds for the Central Universities(No.1082204112364)+4 种基金the Sichuan University Luzhou Municipal Government Strategic Cooperation Project(No.2022CDLZ-8)the Key R&D Program of Sichuan Province(Nos.2022YFN0017 and 2023YFG0019)the Natural Science Foundation of Sichuan(No.2023NSFSC0474)the Tianfiu Yongxing Laboratory Organized Research Project Funding(No.2023CXXM14)the Digital Media Art,Key Laboratory of Sichuan Province,Sichuan Conservatory of Music(No.22DMAKL04).
文摘The pursuit of optimal neural network architectures is foundational to the progression of Neural Architecture Search (NAS). However, the existing NAS methods suffer from the following problem using traditional search strategies, i.e., when facing a large and complex search space, it is difficult to mine more effective architectures within a reasonable time, resulting in inferior search results. This research introduces the Generative Pre-trained Transformer NAS (GPT-NAS), an innovative approach designed to overcome the limitations which are inherent in traditional NAS strategies. This approach improves search efficiency and obtains better architectures by integrating GPT model into the search process. Specifically, we design a reconstruction strategy that utilizes the trained GPT to reorganize the architectures obtained from the search. In addition, to equip the GPT model with the design capabilities of neural architecture, we propose the use of the GPT model for training on a neural architecture dataset. For each architecture, the structural information of its previous layers is utilized to predict the next layer of structure, iteratively traversing the entire architecture. In this way, the GPT model can efficiently learn the key features required for neural architectures. Extensive experimental validation shows that our GPT-NAS approach beats both manually constructed neural architectures and automatically generated architectures by NAS. In addition, we validate the superiority of introducing the GPT model in several ways, and find that the accuracy of the neural architecture on the image dataset obtained from the search after introducing the GPT model is improved by up to about 9%.
基金The National Natural Science Foundation of China(No.41574022)the Postgraduate Research&Practice Innovation Program of Jiangsu Province(No.KYCX17_0150).
文摘To improve the applicability of the global pressure and temperature 2 wet(GPT2w)model in estimating the weighted mean temperature in China and adjacent areas,the error compensation technology based on the neural network was proposed,and a total of 374800 meteorological profiles measured from 2006 to 2015 of 100 radiosonde stations distributed in China and adjacent areas were used to establish an enhanced empirical model for estimating the weighted mean temperature in this region.The data from 2016 to 2018 of the remaining 92 stations in this region was used to test the performance of the proposed model.Results show that the proposed model is about 14.9%better than the GPT2w model and about 7.6%better than the Bevis model with measured surface temperature in accuracy.The performance of the proposed model is significantly improved compared with the GPT2w model not only at different height ranges,but also in different months throughout the year.Moreover,the accuracy of the weighted mean temperature estimation is greatly improved in the northwestern region of China where the radiosonde stations are very rarely distributed.The proposed model shows a great application potential in the nationwide real-time ground-based global navigation satellite system(GNSS)water vapor remote sensing.
