Joleen Liang talks about how to upgrade education with AI adaptive education technology
SHANGHAI, June 5, 2019 /PRNewswire/ -- On April 15th local time, the 2019 O'Reilly AI Conference, one of the world's top AI summits, was held in New York, USA and attended by more than 1,000 attendees. Joleen Liang, partner and Senior Vice President of Squirrel AI Learning by Yixue Group, and Richard Tong, chief architect of Squirrel AI Learning, were invited to the conference, where they got together with famous AI experts and scientific education leaders, including Ion Stoica, a professor at Department of EECS of University of California Berkeley; Roger Chen, Chairman of the O'Reilly AI Conference Project and a co-founder of Omputable; Jian Chang, an advanced algorithm expert of Alibaba Group; and Jeremy Lewi, a co-founder of Google Kubeflow, talking freely about the future of "AI+education" and how AI adaptive educaiton technology could upgrade education.
The O'Reilly AI Conference provides a space where cutting-edge science and new business implementation are combined vehemently with each other. New directions and opportunities for the application of emerging AI technologies in practice were discussed in simple terms at all the previous conferences. AI experts analyzed cases, shared with the others new practice in AI, pointed out new AI applications, addressed the limitations on AI, and predicted an immense change to be caused by AI to the world in the future.
The O'Reilly AI Conference is hosted by O'Reilly Media, Inc., a leading publishing company in the world's UNIX, X and Internet and other open book fields, and a pioneer in online publishing. From The Whole Internet User's Guide and Catalog (a best-seller rated one of the 50 most important books of the 21st century by the New York Public Library) to GNN (the first Internet portal and commercial website) and to WebSite (the first Web server software for desktop PC), the O'Reilly Media, Inc. has been a front-runner in Internet development.
Today, the most popular term in the global scientific and technological community is nothing but "AI" , which is considered a revolutionary technique that may lead to a next major technological breakthrough by tech giants including Google, Microsoft, Apple, IBM and Intel. But for a long time, there has been a lack of a proper medium by which AI can go out of the academic circle into the industrial circle, whereas the O'Reilly AI Conference shoulders the mission of achieving this goal by sharing the most intuitive industrial practical experience. According to the data released by the Qianzhan Industry Research Institute, the global AI market will achieve average compound growth of 50.7% by 2015-2025 and may reach USD36.9 billion by 2025. According to CBInsights, the global AI market reached up to about RMB10.93 billion in 2016. Stimulated by technology demand and driven by mature technology, the market will continue to maintain rapid growth in the next few years.
The O'Reilly AI Conference has achieved remarkable results in all the past years. At the previous conferences, famous AI experts, including Feifei Lee, a professor at Stanford; Yann LeCun, chief scientist of Facebook; Andrew Ng, a co-founder of Coursera and a former chief scientist of Baidu; and Manuela Veloso, Head of the Department of Machine Learning of Carnegie Mellon University (CMU), gave speeches and argued with one another, making a great fire with AI as a little spark.
Joleen Liang, a partner and Senior Vice President of Squirrel AI Learning, was invited to the conference to give a keynote speech, sharing with the other participants how Squirrel AI Learning had brought about a bold change in the education industry with innovative methods and advanced techniques such as MCM, MIBA, deep learning and nanoscale knowledge splitting. The conference was also attended by representatives of internationally renowned high-tech and AI companies, including Intel, Microsoft, Cisco and Dell.
Richard Tong, chief architect of Squirrel AI Learning, also briefed the other participants on problems faced by traditional education and what Squirrel AI Learning had spared no efforts to do in the AI+ education field: One of the most critical problems arising from the traditional education is the lack of high-quality teachers that can meet students' individualized needs. The teaching level is irregular due to the severe shortage of excellent teachers and difference in work experience and ability from teacher to teacher. Moreover, local educational resources are unequally distributed and students differ from one another in learning problem and capacity. All these issues need to be solved urgently using an effective tool.
AI technology, especially the AI adaptive technology, can enable the new generation of teachers to teach students more effectively so as to improve the efficiency of the education industry.
Squirrel AI Learning is the first AI adaptive education brand in the K12 secondary and primary education field of the Asia-Pacific region, and China's first high-tech company with a self-developed AI adaptive learning system for secondary and primary education. Powered by a dozen AI algorithm technologies, including knowledge space theory, Bayesian theory, logistic regression, genetic algorithm and deep learning, the Squirrel AI Learning system can accurately capture students' learning intensity and precisely provide them with a suitable learning path, making it possible to "teach students in accordance with their aptitude".
Here Comes an Era of AI: Internationalized Application of AI
At the O'Reilly AI Conference in New York, heads of many well-known industry magnates gave speeches on how to implement the AI technology from an international perspective, and talked about how to find the pain spots in the market with the AI according to their own specific situation.
Aleksander Mandry, an associate professor of the Department of Computer Science at MIT, pointed out the challenges faced by large-scale applications of AI and solutions. Gadi Singer, Vice President of Intel AI, gave a presentation of the current AI industry standards and hardware development, and how they explored four real AI deployments at Intel.
Tony Jebara, Director of Netflix Machine Learning and an associate professor at the Columbia University, provided detailed information on the method used by Netflix to recommend titles to users and discussed how Netflix integrated causality and fairness into its numerous learning and personalized systems. Steve Re, an associate professor of computer science at the Stanford University, shared with those present his Snorkel system, which can process data faster and more efficiently. He emphasized that for the application of machine learning, model, data and hardware must be combined together. Ruchir Puri, chief scientist of IBM Watson, talked about three aspects of AI automation, including AI design, optimization and correction of other AI systems. Then, he presented IBM Watson's experience in these areas.
Deepashri Varadharajan, chief AI analyst at CBInsights, a world-renowned venture capital research institute, introduced CBInsights' latest research findings about AI start-ups. CB Insights tracked more than 3,000 AI start-ups in over 25 vertical industries. While every vertical industry benefits from deep learning and better hardware processing, it has unique bottlenecks and opportunities. Deepashri Varadharajan discussed the feasibility of promoting AI application in different industries such as healthcare, retail and security, and analyzed emerging business models.
Joleen Liang, partner and Senior Vice President of Squirrel AI Learning, the first AI adaptive education brand in the Asia-Pacific region, gave a keynote speech titled "The Way AI Adaptive Technology Upgrades Education", briefing the participants on how Squirrel AI Learning helped students develop comprehensive learning capacity for what is more important than the mastery of knowledge points, and Squirrel AI Learning's frontier research and explorations carried out in conjunction with the SRI International and other leading international research institutes. They used the MIBA multi-mode comprehensive behavior analysis collector to understand students' expressions and movements to better understand their learning status and objectives. They also developed optimization models, including the deep learning model used by Squirrel AI Learning.
Listening to Joleen's introduction, the participants gained a deeper understanding of the huge change brought to the education industry by innovative approaches and advanced AI techniques, and gave praise to Squirrel AI Learning for its breakthroughs.
After the conference, Joleen Liang accepted an interview with Paco Nathan, Director and Editor of the Learning Group of the O'Reilly AI Conference, and introduced Squirrel AI Learning's development history as well as its far-reaching influence on future education and other industries. Learning from Joleen that Squirrel AI Learning had done a lot for the children in the poor and remote areas of rural China to realize educational fairness and provide students with the best education, Paco Nathan shed tears with excitement.
Richard Tong, chief architect of Squirrel AI Learning, explained the relationship between human learning and machine learning at the workshop, and pointed out how to completely change education and learning with the aid of AI. He began with the necessity of applying AI to education, revealing the significance of AI technology in replacing, enhancing, improving and redefining education. Furthermore, he stressed that the new revolution in education required continuous one-on-one education for students, and that human teachers, who taught courses, should start offering psychological guidance to students. Then, from the perspective of four features, including Squirrel AI Learning's business model, experience in AI implementation, massive data in the learning environment and strong computing power, he explained why Squirrel AI Learning could achieve fast development.
Squirrel AI Learning has created an AI-driven adaptive learning engine. As a combination of thinking model, capacity, methodology (MCM), multidimensional probability knowledge state model (PKS), deep learning technology-based multi-mode comprehensive behavior analysis (MIBA) and nanoscale knowledge component (NKC) as well as other AI techniques, such as reinforced learning, this engine can promote the development of AI+adaptive education to influence future education.
Richard also introduced the structure of the Squirrel AI Learning system, and the AI-related projects implemented in cooperation with the world-famous research institutions SRI International and Carnegie Mellon University (CMU). After Richard finished his technical presentation, many participants formed a queue to learn about Squirrel AI Learning's technical details from him.
Adaptive Education: AI-powered Education
At the O'Reilly AI Conference in New York, Squirrel AI Learning showed its strength and idea of AI+ education to the world, as well as ongoing efforts and remarkable achievements in promoting the fairness and individualization of education.
Rome was not built in a day. Although adaptive education, which originated in the United States, has accurately found the pulse of the educational reform, due to the unsophisticated systemic function and extremely high requirements for institution assessment and teaching & research ability, it has been in a process of slow development in the past. However, AI empowerment has given adaptive education new hope on the eve of the AI reform.
The AI technology-based adaptive learning system is composed of more capacity levels, each of which differs slightly from the others and the difference can be seen clearly. It can dynamically adjust the learning content and path according to students' real-time learning status, build a more complex learning model so as to realize individualized education on a large scale.
Knewton, the world's first AI adaptive education provider, has offered services for 20 million students in North America. RealizeIt, the largest individualized and adaptive learning product provider in the higher education field of the USA, offers more than 40,000 adaptive courses.
As the Chinese pioneer in intelligent adaptive education, Squirrel AI Learning primarily provides AI adaptive learning programs and services for primary and secondary school students. Squirrel AI Learning has successfully developed an advanced algorithm-based intelligent adaptive learning engine with full proprietary intellectual property rights, and the Squirrel AI Learning system can serve as an expert teacher, but its educational efficiency is 5-10 times as high as the traditional one.
Behind student achievements is Squirrel AI Learning's emphasis on realistic educational issues, rigorous attitude towards science, and painstaking efforts on research. Squirrel AI Learning has introduced a large number of scientific and technological talents from the global AI field to develop high-tech products with independent intellectual property rights.
Last year, Prof. Tom Mitchell, the global luminary on machine learning and Dean of School of Computer Science, CMU, which ranks first worldwide in terms of AI, joined the Squirrel AI Learning team as chief AI scientist, causing a sensation in the whole AI circle. Squirrel AI Learning has established Yixue AI Lab with SRI as the primary research partner and Chinese Academy of Sciences (CAS) respectively for discussing together on continuous technological innovation capability in the era of AI. Squirrel AI Learning's strength has also been recognized internationally. Its academic papers and workshop have been accepted by top international academic conferences, such as AIED, CSEDU, IJCAI, AERA and NCME. Also, it's come out in front in the global education field, filling in the gap of domestic AI adaptive education.
Technological innovation and education are of great significance to a country. Squirrel AI Learning has gained recognition from the world by constantly bringing forth the new through the old, making innovations and powering education with technology. We believe that in the future, Squirrel AI Learning will surely play an important role in deepening the reform of the global education industry with AI technology.
View original content to download multimedia:http://www.prnewswire.com/news-releases/squirrel-ai-learning-by-yixue-group-is-invited-to-the-2019-oreilly-ai-conference-in-new-york-city-300877696.html