Posted by Datawhale
2023 White Paper on China's Artificial Intelligence Talent Learning
I Introduction
A few days ago, the "2023 China Artificial Intelligence Talent Learning White Paper" (hereinafter referred to as the "White Paper") compiled by Datawhale, Shanghai Magnolia Open Source and Open Research Institute, Hejing Technology, and Jiangnan University Education Informatization Research Center was officially released on August 24.
Big names in the academic world congratulate the release
2023 is known as the first year of civilianization of AI artificial intelligence, and AI technology has fully penetrated into all aspects of the economy and society. The white paper focuses on the reality of the mismatch between the training of AI talents in colleges and universities and the needs of the industry. Through desktop research, questionnaire surveys, in-depth interviews and other research methods, it explores the needs and challenges of artificial intelligence talent training, and actively explores the role of open source learning ecology in the cultivation of artificial intelligence talents in colleges and universities. The above value path provides a feasible direction for many colleges and universities to improve the artificial intelligence talent training system, and provides innovative ideas for deepening the integration of industry and education, promoting industry-university cooperation and collaborative education, and cultivating high-quality talents that meet the needs of industrial development.
(The full text is 12,000 words in total, this article is an overview, the official account replies to " White Paper ", and the full HD file can be downloaded )
II Shocks and Challenges
The development status of artificial intelligence talents in China
With the acceleration of the era of artificial intelligence, on the one hand, the AI talent gap in my country is getting bigger and bigger. In 2022, the AI talent gap in my country will reach 5 million. Maimai's "2023 Talent Report" shows that the number of AI-related jobs has increased by 40%. The demand/talent delivery is only 0.83; on the other hand, with the rapid change and popularization of artificial intelligence technology, the demand for innovative compound talents in the AI job market will further expand, and the ability requirements for AI talents are also increasing. Continuous improving.
Based on this, the contradiction between the "low employment rate of college graduates and the gap in the demand for industrial talents" in my country has become increasingly prominent, and the development of AI talents is facing problems such as insufficient stock, low quality, and limited growth. Under the impact of demand, the cultivation of artificial intelligence talents in colleges and universities is facing greater challenges.
III Decoupling of production and education
The Pain Points of Artificial Intelligence Talent Training in my country's Colleges and UniversitiesAfter surveying 2,000+ college teachers and students and conducting in-depth interviews with 50+ teachers and students, we found that there are breaks in all aspects of talent cultivation, from knowledge acquisition, to ability training, to scenario application, to innovation challenges, leading to There is no match between cultivating talents and industry needs. In addition, teachers and students generally reflect that the existing talent training system in colleges and universities has problems such as lack of practical opportunities, lack of teacher-student interaction, outdated teaching content, and too theoretical teaching.
IV Diversified accommodation
New Trends of Learning AI Talents in Colleges and Universities
Based on the analysis of survey data, we have summarized four current learning modes: broadcast-style college learning, converging online learning, network-style community learning, and problem-solving practical application learning, and compared and analyzed their advantages. disadvantage. Our research found that in the future, there will be a trend of integration of AI talent education and learning. That is to give full play to and integrate the advantages of different learning methods to meet the individualized learning needs of students at different stages of school learning.
V open source ecology
New ways of learning AI talents in colleges and universities
On this basis, in order to further implement industry-university cooperation and deepen the integration of industry and education, we have discovered a new Internet-based learning method - open source learning, which emphasizes the interaction, sharing and cooperation among learners, and through the joint creation of knowledge , collaborative problem solving and innovative applications to achieve learning goals.
Open-source learning has two main characteristics: strong network properties and strong connection properties. Open source learning has built a diverse learning ecological network, allowing talents to be infinitely close to the industry while realizing the efficient connection of people-knowledge-scenario. In open source learning, students can acquire more cutting-edge industry knowledge, understand actual industry scenarios, improve practical capabilities, and incubate innovative application capabilities.
Open source learning enables learners to upgrade from a single knowledge consumer to a knowledge producer, further enhances learner participation, strengthens learning motivation, and improves learners' sense of gain and accomplishment while exercising their comprehensive abilities.
Based on the research on the learning ecology of open source talents, we divided the elements of the open source ecology into resource type, course type, platform type and community type according to the attributes, forming an artificial intelligence open source learning ecological map.
VI breaks first and then stands
Prospects for Cultivating Artificial Intelligence Talents in Universities
We look forward to the formation of a connected, multi-integrated higher education ecology in the future of artificial intelligence talent training, and based on this, we propose several innovative prospects:
1. Based on the learning platform, break through the "wall" of the classroom and construct a multi-integrated learning space;
2. Keep up with changes in the industry to achieve a dynamic, sustainable and operable supply of resources;
3. Driven by "teaching + task", create an open, connected and project-based training model;
4. Build a result-oriented, process-oriented evaluation mechanism with two-way feedback from teachers and students.
Click "Read the original text" to explore the complete content of the white paper.