=====================================================================================
College of Computing and Mathematics
Computer Engineering Department
=====================================================================================
Presents Public Seminar
"IMCA: An Efficient In-Memory Convolution Accelerator"
Date: Wednesday, March 29, 2023,
Time: 02:30 PM – 03:30 PM
Location: Bldg. 22, Room 132
Speaker:
Mr. Hamed Al-Mehdhar
MS Student, Computer Engineering Department
KFUPM
Abstract:
Traditional convolutional neural network (CNN) architectures suffer from two bottlenecks: computational complexity and memory access cost. In this study, an efficient in-memory convolution accelerator (IMCA) is proposed based on associative in-memory processing to alleviate these two problems directly. In the IMCA, the convolution operations are directly performed inside the memory as in-place operations. The proposed memory computational structure allows for a significant improvement in computational metrics, namely, TOPS/W. Furthermore, due to its unconventional computation style, the IMCA can take advantage of many potential opportunities, such as constant multiplication, bit-level sparsity, and dynamic approximate computing, which, while supported by traditional architectures, require extra overhead to exploit, thus reducing any potential gains. The proposed accelerator architecture exhibits significant efficiency in terms of area and performance, achieving around 0.65 GOPS and 1.64 TOPS/W at 16-bit fixed-point precision with an area of less than 0.25 mm2.
Speaker Bio:
Mr. Hamed Al-Mehdhar is an MSc student in the Computer Engineering Department at KFUPM. He holds a BSc degree in Computer Engineering from KFUPM (2014). Mr. Hamed worked in Sustainable Technology and Engineering Solutions (STES), a local startup specializing in security systems and system integration, as a technical sales engineer and then as a project manager (2014 - 2021). Mr. Hamed is interested in Digital Systems and Design.
All faculty, researchers and graduate students are invited to attend.
=====================================================================================
Computer Engineering Department, College of Computing and Mathematics
Telephone: +966 (13) 860 2110, Email: c-coe@kfupm.edu.sa, Website: www.kfupm.edu.sa/departments/coe/
Copyright © 2014 King Fahd University of Petroleum & Minerals
=====================================================================================
