Computer System Architecture Lab
Computer System Architecture Lab
Home
News
Members
Publications
Research
Gallery
Contact
Light
Dark
Automatic
Accelerator
LogFlex: Flexible-Bit Log Arithmetic Accelerator for Language Models on Edge
Deploying language models (LMs) on resource-constrained mobile/wearable devices while maintaining the output quality is challenging. To …
Yujin Kim
,
Faraz Tahmasebi
,
Gunjae Koo
,
Hyoukjun Kwon
PDF
Cite
Project
DOI
Performance and Cost Analysis of an Embedded Accelerator Platform for CNN Workloads
Modern artificial intelligence (AI) applications operate on a wide range of systems. Researchers have presented hardware and software …
Dowon Kwon
,
Gunjae Koo
PDF
Cite
Project
DOI
FINEA: An Efficient Neural Network Accelerator Exploiting Factorized Input Features
Modern deep neural network (DNN) models increasingly adopt quantized data formats to alleviate the computational burdens of convolution …
Yujin Kim
,
Chanhun Jeong
,
Yunho Oh
,
Myung Kuk Yoon
,
Gunjae Koo
PDF
Cite
Project
Slides
DOI
Slide Show
HALO: Hybrid Systolic Arrays via Logical Partitioning for Acceleration of Complex-Valued Neural Networks
Complex-Valued Neural Networks (CVNNs) are an emerging class of deep learning models that process data with both real and imaginary …
Ji Yeong Yi
,
Eunbi Jeong
,
Sunghee Yum
,
Jane Rhee
,
Sangun Choi
,
Gunjae Koo
,
Yunho Oh
,
Myung Kuk Yoon
PDF
Cite
Project
Slides
DOI
Slide Show
Poster
HyMM: A Hybrid Sparse-Dense Matrix Multiplication Accelerator for GCNs
Graph convolutional networks (GCNs) are emerging neural network models designed to process graph-structured data. Due to massively …
Hunjong Lee
,
Jihun Lee
,
Jaewon Seo
,
Yunho Oh
,
Myung Kuk Yoon
,
Gunjae Koo
PDF
Cite
Project
Slides
DOI
Slide Show
Poster
Performance Analysis of Graph Convolutional Networks on Accelerator Architectures
Graph neural networks의 연산 과정은 높은 희소성을 갖는 행렬 연산을 포함하고 있다. 그러므로, GNN …
Hunjong Lee
,
Gunjae Koo
PDF
Cite
Project
Slides
Slide Show
Audio Compression Accelerator Design for Improving the Response Time of AI Speakers
Response time is one of the critical performance factors of artificial intelligence (AI) speakers. The internet network delays and the …
Hunjong Lee
,
Junhwan Yoo
,
Gunjae Koo
PDF
Cite
Project
Slides
Video
Slide Show
Cite
×