本文へジャンプ

成果報告書詳細
管理番号20190000000263
タイトル*平成30年度中間年報 NEDOプロジェクトを核とした人材育成、産学連携等の総合的展開 実データで学ぶ人工知能講座
公開日2019/6/25
報告書年度2018 - 2018
委託先名国立大学法人東京大学
プロジェクト番号P06046
部署名ロボット・AI部
和文要約
英文要約Title: Educational program to learn Artificial Intelligence with actual data/Comprehensive development of human resources development and industry-academia collaboration centered on NEDO project (FY2017-2019) FY2018 Annual Report

In this project, we develop an educational program of artificial intelligence (AI) to foster human resources sought by industry who possess various knowledge and generalizing abilities related to AI, who can make full use of programming skills and knowledge of computer science (CS), and who can apply his/her skills to individual social tasks. To that end, we build a package that trains human resources in a short term so that they can have practical earning power on the basis of the knowledge and skills of CS and cutting-edge AI technology. The quality is assured by capability evaluation. The package consists of the following four processes: (1) A placement test that measures one’s basic academic knowledge and skills of CS at the time of introduction. (2) Top level lectures for acquiring basic and leading knowledge on AI. (3) Exercises dealing with real common data of concrete social tasks to raise an industry-ready talent. (4) Capability evaluation at the end of the exercises for quality assurance.

In the fiscal year of 2018, we provided a full program twice throughout the year. We continued developing a set of CS placement tests and editing teaching materials that consist of CS supplementation, AI's top-level lectures and AI exercises. CS supplementation covers algorithms and Python programming. For the AI lectures, we added more topics about AI’s history, machine learning basics (regression, cross validation, regularization, trees, etc.), methods related to deep neural networks (fully connected networks, convolutional networks and recurrent networks), search, supervised/unsupervised learning, reinforcement learning, and introduction to computer vision and natural language processing. For the AI exercises, we prepared a set of textbook and test codes about the topics listed above in addition to exercises about image recognition, natural language processing, reinforcement learning and about some libraries. We purchased one server and managed the four servers (three were purchased last year) and a website for the program.

For the summer and winter terms, we received 131 and 167 applications respectively, and recruited 46 and 53 students (from 41 and 38 companies) after screening. The program was conducted for 14 days in total in summer, and 13 days in total in winter, where three days were for CS supplementation, and 11 (in summer) and 13 days (in winter) were for AI lectures and exercises. We also had lunch meeting to facilitate communication between the students and the lecturers. We had poster presentations from the students on the last day of the program and invited lecturers from Osaka university to evaluate the presentations. 44 and 43 students have completed the program and received the certifications.
ダウンロード成果報告書データベース(ユーザ登録必須)から、ダウンロードしてください。

▲トップに戻る