A 1ynm 1.25V 8Gb, 16Gb/s/pin GDDR6-based Accelerator-in-Memory supporting 1TFLOPS MAC Operation and Various Activation Functions for Deep-Learning Applications-Reference-Cited by-同舟云学术

A 1ynm 1.25V 8Gb, 16Gb/s/pin GDDR6-based Accelerator-in-Memory supporting 1TFLOPS MAC Operation and Various Activation Functions for Deep-Learning Applications

Published:2022-02-20 Issue: Volume: Page:
ISSN:
Container-title:2022 IEEE International Solid- State Circuits Conference (ISSCC)
language:
Short-container-title:

Author:

Lee Seongju¹,Kim Kyuyoung¹,Oh Sanghoon¹,Park Joonhong¹,Hong Gimoon¹,Ka Dongyoon¹,Hwang Kyudong¹,Park Jeongje¹,Kang Kyeongpil¹,Kim Jungyeon¹,Jeon Junyeol¹,Kim Nahsung¹,Kwon Yongkee¹,Vladimir Kornijcuk¹,Shin Woojae¹,Won Jongsoon¹,Lee Minkyu¹,Joo Hyunha¹,Choi Haerang¹,Lee Jaewook¹,Ko Donguc¹,Jun Younggun¹,Cho Keewon¹,Kim Ilwoong¹,Song Choungki¹,Jeong Chunseok¹,Kwon Daehan¹,Jang Jieun¹,Park Il¹,Chun Junhyun¹,Cho Joohwan¹

Affiliation:

1. SK hynix,Icheon,Korea

Publisher

IEEE

Link

http://xplorestaging.ieee.org/ielx7/9731529/9731102/09731711.pdf?arnumber=9731711

Reference7 articles.

1. Newton: A DRAM-maker’s Accelerator-in-Memory (AiM) Architecture for Machine Learning

2. 25.4 A 20nm 6GB Function-In-Memory DRAM, Based on HBM2 with a 1.2TFLOPS Programmable Computing Unit Using Bank-Level Parallelism, for Machine Learning Applications

3. A study of BFLOAT16 for deep learning training;kalamkar;arXiv (pre-ottntt id 1905 12322,2019

4. 25.3 An 8Gb GDDR6X DRAM Achieving 22Gb/s/pin with Single-Ended PAM4 Signaling

5. Roles and Missions for National Changgeuk Company of Korea to Eatablish Universal Korean Music Theatres: Mainly on Changgeuks, Chung and Romeo And Juliet

Cited by 19 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. A Floating-Point 6T SRAM In-Memory-Compute Macro Using Hybrid-Domain Structure for Advanced AI Edge Chips;IEEE Journal of Solid-State Circuits;2024-01

2. In-memory computing technology: development status, trends and challenges;SCIENTIA SINICA Informationis;2023-12-26

3. GEBA: Gradient-Error-Based Approximation of Activation Functions;IEEE Journal on Emerging and Selected Topics in Circuits and Systems;2023-12

4. AESPA: Asynchronous Execution Scheme to Exploit Bank-Level Parallelism of Processing-in-Memory;56th Annual IEEE/ACM International Symposium on Microarchitecture;2023-10-28

5. SimplePIM: A Software Framework for Productive and Efficient Processing-in-Memory;2023 32nd International Conference on Parallel Architectures and Compilation Techniques (PACT);2023-10-21