スパイク発生分子素子を利用した確率的コンピューティン物理実装とそのAI応用

2023 Fiscal Year Annual Research Report

• Project/Area Number: 22KJ0091
• Allocation Type: Multi-year Fund
• Research Institution: Hokkaido University
• Principal Investigator: KAN SHAOHUA 北海道大学, 情報科学院, 特別研究員(PD)
• Project Period (FY): 2023-03-08 – 2024-03-31
• Keywords: in-material computing / information processing

Outline of Annual Research Achievements

My research achievements were mainly did at the Nano-Electronics of the University of Twente in the Netherlands this year. I investigated information processing using atom-doped silicon-based nanodevices. By improving the output circuit, the short term memory ability is introduced into the device, and excellent results are obtained in the feature extraction of gesture video signals by applying these techniques to the processing of time signals. Based on these results, I won the Excellent Poster Award at the In-material computing workshop for young researchers hold in Sapporo.

Research Products

• [Int'l Joint Research] トゥウェンテ大学(オランダ)
  • Country Name: NETHERLANDS
  • Counterpart Institution: トゥウェンテ大学
• [Presentation] Dopant-atom network processing unit serves as convolution kernels (2023)
  • Author(s): Shaohua Kan
  • Organizer: インマテリアルコンピューティング若手研究会
  • Int'l Joint Research / Invited
• [Funded Workshop] In-material computing workshop for young researchers (2023)