Squirrel AI Learning Shines at the IEEE/IROS 2019 Conference: The Era of AI + Education has Arrived

Squirrel AI Learning Shines at the IEEE/IROS 2019 Conference: The Era of AI + Education has Arrived

MACAO, Dec. 17, 2019 /PRNewswire/ -- The 32nd IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2019) was held in Macao from November 4 to November 8, 2019. IROS is the flagship international conference on robotics and intelligent systems. It is one of the two top international conferences in the fields of intelligent robotics and automation, and also the most influential international top robot academic conference in the world.

Dr. Wei Cui, Chief Scientist of Squirrel AI Learning, was invited to attend the IROS 2019 conference and delivered a wonderful speech. Speakers on the same stage also included Toshio Fukuda, General Chairman of IEEE2020, Founder of IROS, Kristen Grauman, AI Research Scientist of Facebook, Professor of Computer Science Department at the University of Texas at Austin, etc.

At the conference, he shared with the top scholars all over the world the technical advantages and achievements in technology driven landing application of Squirrel AI Learning intelligent adaptive teaching system, showed everyone the outstanding achievements made by Chinese enterprises in the field of AI adaptive education, and also allowed the participants know about the development and application of artificial intelligence in the field of education, which attracted great attention from famous experts and scholars in the international industry.

This conference is jointly organized by IEEE, IEEE Robotics and Automation Society (RAS) and IEEE Industrial Electronics Society (IES), etc. It has attracted more than 4,000 professionals and representatives of top research teams in artificial intelligence and other fields around the world.

Dr. Wei Cui, Chief Scientist of Squirrel AI Learning, Delivered a Speech
Dr. Wei Cui, Chief Scientist of Squirrel AI Learning, Delivered a Speech

Squirrel AI Learning is the first artificial intelligence company in China to apply AI adaptive learning technology to the field of K12 education. Squirrel AI Learning has successfully developed the first AI adaptive learning engine with complete independent intellectual property rights and advanced algorithms as the core in China.

Squirrel AI Learning intelligent adaptive learning system is a student-centered intelligent and personalized education, it applies artificial intelligence technology in the process of assessment, learning, practicing, testing and questioning, to achieve the purpose of surpassing the human teaching on the basis of simulating excellent teachers. The product has a high cost-effectiveness performance. It adopts the mode of artificial intelligence + human teachers to teach students according to their aptitude, which can effectively solve the problems of high class cost, few famous teacher resources and low learning efficiency of traditional education.

At the conference, Dr. Wei Cui said: "Artificial intelligence will bring personalized education to every student. Through technical means, we hope that every student in China can enjoy the charm of personalized education that taught according to their aptitude."

IROS 2019 is jointly organized by IEEE, IEEE Robotics and Automation Society (RAS), IEEE Industrial Electronics Society (IES), Robotics Society of Japan (RSJ), Society of Instrument and Control Engineers (SICE) and New Technology Foundation (NTF).

The conference attracted more than 4,000 professionals, representatives of top research teams and business people from all over the world in robotics, automation systems, artificial intelligence and other fields to participate in the exchange, jointly explore the cutting-edge technologies in the field of intelligent robots and systems, share and discuss the latest progress in related fields.

Toshio Fukuda, General Chairman of IEEE2020, Founder of IROS
Toshio Fukuda, General Chairman of IEEE2020, Founder of IROS

Dr. Wei Cui, who was invited to attend the IROS 2019 conference, is the Chief Scientist of Squirrel AI Learning, and is also the earliest promoter of adaptive education in China. He was selected as the winner of MIT Technology Review 35 under 35 by MIT Technology Review, and is a recognized senior engineer with Shanghai AI senior professional title in 2019, which can be described as a young talent. Through his own efforts, Dr. Wei Cui is working hard to make AI technology change the pattern of education development in China.

Squirrel AI Learning: Intelligent Adaptive Education Technology Changes Education Pattern

At this IROS 2019 conference, Dr. Wei Cui introduced in detail to the participants how the Squirrel AI Learning intelligent adaptive education system can make its own contribution to the industrialization of education through the empowerment of AI technology.

At the technical level, Squirrel AI Learning fully uses more than ten kinds of artificial intelligence algorithm technologies, such as knowledge space theory, Bayesian theory, logistic regression, genetic algorithms and deep learning, and initiates the nanoscale knowledge point decomposition, the decomposition of MCM system (Model of Thinking, Capacity and Methodology), and the deep decomposition of knowledge points, clearly and accurately penetrates the students' knowledge loopholes/weaknesses; the nanoscale knowledge point splitting created originally by Squirrel AI Learning splits the knowledge points in the disciplines into nanoscale knowledge point splitting.

Through the correlation between knowledge points, students' user portraits can be visualized, so as to have a clearer understanding on the grasp of students' knowledge points and accurately detect the weak points of students' knowledge points. On the basis of nanoscale knowledge point decomposition, the learning abilities and learning methods of Squirrel AI Learning for students are split into measurable and teachable learning ability models.

On this basis, Squirrel AI Learning intelligent adaptive learning system can accurately detect the weak points of learning knowledge, accurately give the most suitable learning path for each child, so as to achieve personalized learning program. The association concept algorithms of non-associated knowledge points created by Squirrel AI Learning not only establishes the relevance of knowledge points based on the knowledge map theory, tracing the source to present, but also establishes the association probability on the non-associated knowledge points, so that the test efficiency and learning efficiency can be improved by 3-10 times respectively than the knowledge map theoretical model of similar products.

It is worth mentioning that Squirrel AI Learning is the first one in the world to put forward the concept of reconstructing knowledge map of fault causes. It sums up the mistakes of every student's knowledge points so as to find the root of problems. And this is also the basis of achieving a real personalized learning program.

In the past, in order to cover the blind area of knowledge points, students often had to carry out the excessive assignments tactics to eliminate illiteracy. Squirrel AI Learning scans the students' knowledge map to accurately locate the blind area of their knowledge. Students do not need to consolidate their knowledge in hundreds of thousands of exam questions. The system will help students quickly according to their weak points of knowledge, so that students' learning efficiency can be maximized, and their performance can be ensured to be improved steadily.

Dr. Wei Cui introduced that at present, in terms of technical characteristics, Squirrel AI Learning is mainly applied to several major AI technologies:

First, evolutionary algorithms, logistic regression and neural networks. The algorithm model takes into account the learning objectives and knowledge state of students, and dynamically adjusts the learning path. The system will gradually draw students' learning habits, interests, methods and other multi-directional student portraits, and continuously optimize the content recommendation logic automatically.

Second, machine learning, deep learning and natural language processing technology. The technology can recommend the most suitable learning content according to different students' personal preferences, learning habits and learning styles. Different students have different degrees of adaptation to different learning atmospheres and difficulties, using natural language processing technology to automatically generate learning content labels; using deep learning technology to analyze student portrait and learning content, and automatically select the suitable learning content for students.

Third, Bayesian theory and Bayesian networks. Look at problems dynamically based on experience and information. The system will render the knowledge points and probability distribution, and predict the learning ability of learners. By comparing the correlation between different knowledge points and the learning degree of students on knowledge points, Squirrel AI Learning intelligent adaptive learning system can infer the correlation between knowledge points without logical relationship.

Fourth, graph theory, knowledge space theory and information entropy theory. Squirrel AI Learning distinguishes between knowledge points according to difficulty, importance and cognitive level, models knowledge system, builds "Knowledge Map", and combs the logic and cognitive correlations between knowledge points.

From the perspective of measurement, information can be quantified. By using the theory of information entropy, we can quickly approach the level of students' knowledge state by detecting some important knowledge points, and then make repeated refined calculations around this basic level, so as to effectively and accurately diagnose students' knowledge loopholes and states.

Fifth, knowledge tracing theory. The system will evaluate the ability level of students at each knowledge point, and analyze the ability level of knowledge points and related knowledge points successively, and finally accurate to the grasp of each nanoscale knowledge point, and update the ability value changes of students in real time after learning, so as to accurately push the most suitable learning path and learning content for the current situation of students. It can not only understand the degree of mastery of students' current knowledge points, but also reflect students' potentiality, which is a kind of prediction.

Sixth, education data mining and learning analysis technology. Education data mining refers to the quantitative analysis of learning process and learning behavior, and collect the learning data of students in the process of learning, including learning time, residence time and test accuracy, etc. Through the processing and analysis of data, different students' learning models are established. Learning analysis technology is mainly used to predict and monitor students' test results, which can provide teachers with detailed student data and information for the system and teachers to improve teaching methods.

Seventh, it is under research that user interaction driven by Dialog-based HUI and VPA engine in the form of dialogue realizes real-time voice interaction between virtual teachers and students. The main technologies are natural language processing, voice recognition and semantic analysis. In the process of learning, students can ask virtual teachers about their learning situation and learning tasks at any time, and give feedback on questions.

Eighth, MIBA is another project working with both SRI and CMU. MIBA refers to the Multi-modal Integrated Behavior Analysis. It collects students' physiological data and behavioral data through cameras, brainwaves measuring rings and other devices, including facial expression data, blood change data under the skin, body movement data and brain wave data. Combined with the learning data generated during the learning process, it analyzes the learning state of students, including the degree of students' concentration and learning input. The teacher terminal system used by teachers can receive early warning signals, and can timely implement personalized intervention to make learning more effective.

Dr. Wei Cui said that just because of the integration of technology, at present, Squirrel AI Learning has established laboratories in cooperation with several top AI research institutes around the world such as CMU, SRI, etc. Squirrel AI Learning intelligent adaptive learning system has opened more than 2,300 learning centers in more than 700 cities and counties in more than 20 provinces, with a total of nearly 2 million registered students.

After understanding the practical application effect of leading algorithms, model and intelligent adaptation technology of Squirrel AI Learning, experts and scholars and participants on site highly praised the achievements of Squirrel AI Learning in research and development and the application of AI technology in the field of education, as well as the behavior that it is committed to bringing fair, efficient and personalized education to every child.

At last, Dr. Wei Cui said that traditional education currently faces many deep-rooted problems: the development of education resources is uneven, high-quality education resources tend to developed areas, while in the underdeveloped areas, the quality of teachers is uneven. Squirrel AI Learning by Yixue Group hopes to achieve the education concept that we have been advocating since ancient times through the empowerment of AI technology: teaching students according to their aptitude, to reform and innovate traditional education.

Cision View original content to download multimedia:http://www.prnewswire.com/news-releases/squirrel-ai-learning-shines-at-the-ieeeiros-2019-conference-the-era-of-ai--education-has-arrived-300975855.html

This website is best viewed using the latest versions of web browsers.