Self-learnable Analog-Digital-Mixed VLSI Processors for Smart Human-Computer-Interaction

2017 Fiscal Year Final Research Report

• Project/Area Number: 26870227
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Human interface and interaction; Electron device/Electronic equipment
• Research Institution: Nara Institute of Science and Technology (2017); Japan Advanced Institute of Science and Technology (2014-2016)
• Principal Investigator: ZHANG Renyuan 奈良先端科学技術大学院大学, 情報科学研究科, 助教 (00709131)
• Project Period (FY): 2014-04-01 – 2018-03-31
• Keywords: Analog-Digital-Hybrid / Approximate Computing / Analog Calculator / Multi-Valued-Logic

Outline of Final Research Achievements

The analog-digital-hybrid computing processors have been developed for the human-computer-interaction (HCI) relevant applications. The approximate computing technologies were implemented in silicon to offer a different computing methodology from traditional binary processing on the basis of “deductive computation”. In this manner, the performances on computing speed and energy consumption were greatly improved. The ACU processor was successfully developed as the world-first programmable analog calculator for multi-operand calculation, which greatly speeds up the widely applied HCI technologies such as CNN.

Free Research Field

Computer Architecture