A Highly Efficient STT-MRAM-Based IndRNN Accelerator with Approximate Dictionary Encoded Weight for Edge AI Devices

Yiming Song

doi:https://doi.org/10.7567/SSDM.2023.J-5-04

Presentation information

Oral Presentation

12: Advanced Circuits / Systems Interacting with Innovative Devices and Materials

[J-5] AI and Neural Applications

Fri. Sep 8, 2023 9:00 AM - 10:15 AM Room J (233, Bldg. 2)

Session Chairs: Koh Johguchi (Shinshu Univ.), Yitao Ma (Zhejiang Univ.)

10:00 AM - 10:15 AM

[J-5-04 (Late News)] A Highly Efficient STT-MRAM-Based IndRNN Accelerator with Approximate Dictionary Encoded Weight for Edge AI Devices

^○Yiming Song¹, Tao Li¹, Tetsuo Endoh¹ (1. Tohoku Univ. (Japan))

https://doi.org/10.7567/SSDM.2023.J-5-04

Abstract password authentication.
PA password is required to view abstracts. You can find the password in the "Advance Program".The "Advance Program" is handed out at the registration desk to registered participants.

Presentation information

[J-5] AI and Neural Applications

[J-5-04 (Late News)] A Highly Efficient STT-MRAM-Based IndRNN Accelerator with Approximate Dictionary Encoded Weight for Edge AI Devices

Password