ai发展意见英语作文

【#第一文档网# 导语】以下是®第一文档网的小编为您整理的《ai发展意见英语作文》，欢迎阅读！

ai发展意见英语作文



First, it is suggested to pool all forces to build a new urban infrastructure represented by AI computing centers, so that AI computing power can become a new urban public resource like water and electricity, and that computing power can become a widely accessible and affordable public resource.

second, only, Hu Houkun also reminded attention to reasonable planning, "The planning of an AI computing center should be coordinated where computing power is concentrated, Instead of casting the net, Especially to avoid repetitive construction; at the same time, It is suggested that the planning work should be focused on greatly, For architectural considerations ending, And deployment and construction should start small, From near to far away, Follow the change in demand, Small steps and fast running, continuous iteration and upgrade; More importantly, In this process, Government technology providers and the majority of application units should work together, While building and guiding the application of the landing, Only in this way can the whole local


industrial upgrading and development be promoted.”

Second, it is suggested that technology should be deeply rooted, "especially for processors, AI computing frameworks, to enhance the resilience of the AI technology industry," Hu said.

Third, it is suggested to boldly use technical means to break through the bottleneck of the current AI universal benefit."I think the bottleneck in the popularization of AI is not in technology, the technology has developed quite well, and it is not in application, because the demand for applications has been booming.Its bottleneck lies in the efficiency of development, now the development efficiency is too low, the development of AI applications is too slow, it seriously hinders the combination of technology and demand.”

Investigate its reason, Hu Houkun said that the current AI development mode is more traditional, need highly dependent on experts and data, not only need to spend a lot of time to collect data, processing data, also need to spend a lot of manpower for parameter tuning, it is difficult to meet the product requirements in a short time, " this mode is called manual workshop mode, we


should use technical means to change this way, improve the efficiency of development.In our view, with large scale training model can solve the problem, based on the training model, developers can only need a small amount of industry data can quickly develop higher accuracy, generalization ability stronger AI model, we call this model'industrial AI development new model, we think this mode of transformation, will greatly promote the development of AI industry.”

本文来源：https://www.dy1993.cn/RKr.html