2021 KIOXIA EVENT ACTIVITY REPORT

Keio AIC and Kioxia collaborated  to hold an idea contest for five weeks from October 7 to November 4, 2021.

During the Contest

In this contest, participants deepened their knowledge of AI through a series of lectures and group work sessions.

In the lectures, participants learned about AI technologies in the world and their applications and effectiveness by referring to the AI Map (*1). They also learned about pseudo-image generation and how to create a blueprint in order to obtain a desired shape. This helped participants further their understanding of semiconductor design technology.

During group work, participants were asked to think about real-life issues and to come up with a plan to use AI to solve them. On the final day of the contest, each group presented their plan.

The results of the contest and details of the participating teams are as follows.

Contest Results

Winner: Team B

Third year student, Faculty of Science and Technology, Hiroshi Omich
First year student, Faculty of Science and Technology, Asahi Inoue

Theme: AI Triage

Overview: Triage is the process of prioritizing people to be treated and selecting hospitals in situations where the number of people that can be treated is limited, such as during a disaster. Currently, triage requires a decision to be made within 30 seconds and is only 70% accurate; AI can help triage decisions be more accurate and thus save more lives.

Evaluation: Triage is an ethically challenging issue, but the problem was clearly identified and numerical targets were set. It was good that the small details of the AI Map were considered as well.

Runner-up: Team F

Doctoral second year student, Graduate School of Media and Governance, Motofumi Saito
Third year student, Faculty of Pharmacy, Mirei Kimura
Second year student, Faculty of Economics, Masanobu Nagasaki

Theme: Mental Doctor MAI

Overview: There are many cases of suicide in Japan due to mental illness. However, mental disorders are very difficult to diagnose because there are many different types of disorders and their symptoms are very similar to that of poor nutrition and sleep deprivation. Therefore, AI can help make accurate diagnoses by analyzing data from patients’ social media and data obtained during medical examinations. Furthermore, by having AI constantly analyze an individual’s personal data, we hope that people will be able to lead comfortable lives with a peace of mind and sense of security, knowing that their well-being is carefully monitored with the help of AI

Evaluation: I felt that the idea of AI providing a sense of security was a challenging theme. I also liked the fact that the proposal even included a vision for the future.

Presentation Award: Team E

Doctoral second year student, Graduate School of Media Design, Kayoko Nakamura
Second year student, Faculty of Science and Technology, Hiroki Nakajima
Second year student, Faculty of Economics, Akine Akasaka
First year student, Faculty of Science and Technology, Kazumitsu Fukuda

Theme: Complete Theme Park Strategy

Overview: One of Japan’s leading theme parks has many attractions, and there are problems such as visitors not knowing the order in which they should visit the attractions and overcrowding at certain attractions. To solve these problems, AI can suggest routes depending on the situation and number of members of the group at that time. This minimizes travel time and waiting time while maximizing user satisfaction.

Evaluation: The ability to forecast wait times in combination with existing applications that display wait times for attractions was interesting. Also, the Q&A was easy to understand because it was prepared in advance.

Team A and Team D

Third year student, Faculty of Economics, Hiroki Fujii
Second year student, Master’s program, Graduate School of Health Management, Hiraga
Third year student, Faculty of Science and Technology, Kota Okudo

Theme: Well-being Improvement Apps

Overview: Many people suffer from mental health issues due to the modern tendency toward individualism and the reduced social interaction in their lives due to the COVID-19 pandemic. However, it is difficult for them to go to the hospital because they are afraid of the risk of COVID-19 infection. Therefore, we came up with an application that collects detailed information about the user’s life, predicts the person’s mental state using AI, and produces optimized data. For example, when the user is sad, it plays music that lifts his/her mood. This makes it possible to prevent a mental disorder from worsening, rather than cure it.

Evaluation: I felt that the key point is that individuals can use it comfortably. Also, I enjoyed the dynamic structure of the AI Map.

Team C

Third year student, Faculty of Science and Technology, Yuto Shiot
Second year student, Faculty of Low, S. Miyauchi
Anonymous

Theme: Course Selection Advice

Overview: When selecting courses, it is very difficult to decide which courses are the most suitable, since the student needs to take into account the number of required courses and the time required for part-time work and club activities. To solve this problem, we came up with a system in which students enter their course requirements and areas of interest, and the AI suggests courses that suit them. The key points are that the AI only provides advice, with the students having to make their own decisions, and that the data on students’ courses of study can be used for job hunting or to find partners for business startups.

Evaluation: Although the theme was simple and familiar to the students, this idea could be modified and used for various purposes, such as job hunting.

Participants’ Comments (Excerpts)

• I learned a lot and had a lot of fun by experiencing the whole process!
• I was able to hear many different opinions and expand my horizons.
• I discovered many different aspects of AI. I was also impressed by the fact that the focus is on people rather than the latest technology.
• I am not very good at participating in discussions, but thanks to Kioxia’s staff, we were able to have discussions in an atmosphere that was not too formal, and I was able to experience the fun and enjoyment of deepening discussions and seeking solutions to problems.

At the end of the contest, a group photo was taken with all who participated in the award ceremony, including the employees from Kioxia, Keio AIC representatives, and students.

Thank you to all the Kioxia employees who, in addition to planning the event, served as facilitators to stimulate discussion during the event, to everyone who planned and managed the contest, and to the students who participated in the event to the very end!

The winning teams will receive certificates and prizes.

Acknowledgements

We would like to express our sincere gratitude to Tsuyoshi Ito, Osamu Torii, Shinji Takihi, Manpei Karigita, Hiroyuki Yamada, Akira Shimoyama, Ichiro Takahashi, and many others at Kioxia Corporation for their generous support.

Keio AIC Coordinators
Takahiro Yakoh, Kenji Yasuoka, Shigeki Ishikawa, Hiroyuki Ito, Hizuru Ikawa, Mayuko Takesako

Student Administrators
Tatsunari Hocchi, Shota Hashimoto, Masayuki Iwase

(*1): Citation: Society for Artificial Intelligence, “AI Map Beta 2.0 (June 2020 Edition)”

https://www.ai-gakkai.or.jp/resource/aimap