I am a Committee Member of
Extenics Society, Chinese Association for Artificial Intelligence (CAAI) (中國人工智能學會可拓學專業委員會委員) and attended the 1st National
Primary and Secondary School Teacher Training Course for Artificial
Intelligence (首期全國中小學人工智能師資培訓課程) organized by Primary and Secondary School
Working Committee, CAAI (中國人工智能學會中小學工作委員會) and School of Artificial Intelligence,
University of Chinese Academy of Sciences (中國科學院人工智能學院) from 19th to 21st
July 2019 in The High School Affiliated to Renmin University of China (RDFZ) (中國人民大學附屬中學) in Beijing. Group photo was took after the officer speech
finished.
Day 1 (20190719):
I arrived the The High School
Affiliated to Renmin University of China and took a photo for memory. I met a classmate Mr. Bingo Wong (華南師範大學附屬中學黃秉剛) who is from Guangzhou.
So many participants attended the training. Then Mr. Yuan Zhongguo (中國人民大學附屬中學信息技術教研組長袁中果) chaired the opening and Ms. Wang Weining (中國人工智能學會秘書長,北京郵電大學王衛寧研究員) gave welcome speech.
In the beginning, Mr. Ren
Changshan (教育部科技司信息化處任昌山副處長) was the first speaker who explained the “Education
Policy of AI”. He briefed the missions
included optimization of AI innovation system in higher education, talent
training system and AI commercialization and demonstration.
Then Mr. Ren briefed the 10
strategies in education and focused on enhancing Information era education
revolution.
Finally, he mentioned the
perspective on China education development on intelligence era. And quoted the Beijing
Consensus (北京共識) in the International
Conference on Artificial Intelligence and Education (國際人工智能與教育大會) as conclusion.
The second speaker was Prof. Tao
Jianhua (中國科學院自動化研究所所長助理兼任中國科學院大學人工智能學院院長助理、模式識別國家重點實驗室副主任陶建華研究員) and his presentation entitled “Artificial
Intelligence State and Challenge”.
Firstly, Prof. Tao briefed what is AI that is to study and develop the
theory, methodology and technology as well as application in simulating and
extending of human intelligence.
Then Prof. Tao mentioned the three key school of AI and they were symbolicism (符號主義), connectionism (聯結主義) and behavioralism (行為主義). Then the history of AI development was briefed and we were in the 3rd wave.
After that Prof. Tao stated the AI from digital world to physical world and from non-critical to critical impact. Now, it used “AI+X” model to integrate into different industries such as medical, advanced manufacturing, biotechnology, machine-human interaction and culture.
And then he mentioned several
bottom neck of AI such as energy consumption, lack of common sense, motion
control, etc. Finally, he stated the new
generation of AI and AI2.0 was the new direction of AI development. He expected some perspectives in the future including
human brain cognition neutral model, IoT self-cognition operation system, and
super ANN computing and smart robot. Lastly,
he said that the key of AI development was talent so that we needed to invest
more in AI education.
The third speaker was Prof. Zhang
Jiajun (張家俊, 中國科學院自動化研究所, 中國科學院大學人工智能學院) and his topic named “Natural
Language Processing (NLP) and Machine Translation – Past, Current and Future”. NLP is using computer to understand different
human languages that could be read and wrote.
Then Prof. Zhang introduced some
challenges on NLP such as large ambiguity in our language and large unknown
language. The NLP history were discussed from 1946 to 2015.
Prof. Zhang said there were two
key methods to handle it and they were logic-based approach (using rules) and
statistic-based approach (using data). It could be integrated to use both
approach.
Finally, he discussed the NLP development. Currently, the NLP function was good and
applied widely but still had many problems. However, the following four
problems were identified.
i)
Lack of explanation and robust (魯棒性)
ii)
Lack of domain knowledge guide
iii)
Lack of common sense and reasoning
iv)
Lack of memory and discrimination ability
At the end, Prof. Zhang discussed
the future NLP that could be based on brain decoding and knowledge based on
lifelong learning.
During the lunch break, we had lunch
in student canteen.
Very nice and health lunch set
They used automatic payment
system for different lunch set. Each disc has RFID for counting price I think.
Prof. Liu mentioned that ANN is
connectionism (聯結主義). Then he asked two critical
questions that were network structure and learning algorithm included supervised
learning, unsupervised learning and reinforcement learning. And then he briefed
the history of ANN from network structure and algorithm since 1943.
Lastly, Prof. Liu discussed the
future development of ANN would be from small sample learning that included Finetune
from characteristic study and Meta Learning (learning to learn), as well as,
visualization (e.g. Heatmap, Image patch based, Activation map / matrix).
The fifth speaker was Prof. Fan
Lei (首都師范大學教育學院樊磊)
and his presentation was “Implementation Problem of the Course on Information
Technology / Artificial Intelligence”. Firstly, Prof. Fan discussed the content
of AI + Education. They were:
i)
AI Generalization,
ii)
Education digitalization to AI,
iii)
AI in Education System, and
iv)
AI and Science Education Integration.
Then Prof. Fan explained why we
needed AI course because of training our next generation to adopt smart
society. And then he introduced the
course requirement standard such as understanding AI development history,
description of AI algorithm realization process, AI application and practical
experience, etc. The following diagram is
reference from USA.
After that Prof. Fan suggested AI
course with key core in algorithm, programming and data, as well as, design
principles in follows:
i)
Systematic and state-of-the-art AI technologies
ii)
Focused and visualized
iii)
Project oriented and experience by design
iv)
Real data and real life case
v)
Smart society and responsibility
At the end, he quoted a good news
that top university was considered the students who were outstanding in AI.
The last speaker was Mr. Yuan
Zhongguo (中國人民大學附屬中學信息技術教研組長袁中果) and his topic named “The AI Education
Implementation Report of the High School Affiliated to Renmin University of
China (RDFZ)”. He quoted the International
Conference on Artificial Intelligence and Education on 16-18 May 2019 that AI
course establishment in high school and RDFZ established a pilot class.
Then Mr. Yuan introduced their IT
teachers’ background. It observed very
high education background and upto 53% were doctorate degree.
The AI course design example was
showed.
Finally, he introduced the High
School AI training platform under Primary and Secondary School Working
Committee, CAAI.
Reference:
习近平向国际人工智能与教育大会致贺信 - http://www.cicpa.org.cn/Column/hyxxhckzl/zcyxs/201905/W020190524342822810580.pdf
20190719:
The 1st National Primary and Secondary School Teacher
Training Course for Artificial Intelligence (首期全國中小學人工智能師資培訓課程) (Day 1) - https://qualityalchemist.blogspot.com/2019/07/the-1st-national-primary-and-secondary.html
20190720:
The 1st National Primary and Secondary School Teacher
Training Course for Artificial Intelligence (首期全國中小學人工智能師資培訓課程) (Day 2) - https://qualityalchemist.blogspot.com/2019/07/the-1st-national-primary-and-secondary_20.html
20190721:
The 1st National Primary and Secondary School Teacher
Training Course for Artificial Intelligence (首期全國中小學人工智能師資培訓課程) (Day 3) - https://qualityalchemist.blogspot.com/2019/07/the-1st-national-primary-and-secondary_21.html
沒有留言:
發佈留言