Energy-Efficient DNN Training Processors on Micro-AI Systems von Han, Donghyeon, Kang, Sanghoon, Kim, Sangyeob, Lee, Juhyoung, Yoo, Hoi-Jun

Volltext

An Ultra-low-power Mixed-mode Face Recognition Processor for Always-on User Authentication in Mobile Device von Kim, Ji-Hoon, Kim, Changhyeon, Kim, Kwantae, Lee, Juhyoung, Yoo, Hoi-Jun, Kim, Joo-Young

Volltext

A 99.4 fps Optical Flow Estimation Processor with Image Tiling for Action Recognition in Mobile Devices von Lee, Juhyoung, Choi, Sungpill, Lee, Jinmook, Kang, Sanghoon, Yoo, Hoi-Jun

Volltext

Deep Reinforcement Learning Processor Design for Mobile Applications von Lee, Juhyoung, Yoo, Hoi-Jun

Volltext

HNPU: An Adaptive DNN Training Processor Utilizing Stochastic Dynamic Fixed-Point and Active Bit-Precision Searching von Han, Donghyeon, Im, Dongseok, Park, Gwangtae, Kim, Youngwoo, Song, Seokchan, Lee, Juhyoung, Yoo, Hoi-Jun

Volltext

Z-PIM: A Sparsity-Aware Processing-in-Memory Architecture With Fully Variable Weight Bit-Precision for Energy-Efficient Deep Neural Networks von Kim, Ji-Hoon, Lee, Juhyoung, Lee, Jinsu, Heo, Jaehoon, Kim, Joo-Young

Volltext

SNPU: An Energy-Efficient Spike Domain Deep-Neural-Network Processor With Two-Step Spike Encoding and Shift-and-Accumulation Unit von Kim, Sangyeob, Kim, Sangjin, Um, Soyeon, Kim, Soyeon, Lee, Juhyoung, Yoo, Hoi-Jun

Volltext

GANPU: An Energy-Efficient Multi-DNN Training Processor for GANs With Speculative Dual-Sparsity Exploitation von Kang, Sanghoon, Han, Donghyeon, Lee, Juhyoung, Im, Dongseok, Kim, Sangyeob, Kim, Soyeon, Ryu, Junha, Yoo, Hoi-Jun

Volltext

A Low-Power Graph Convolutional Network Processor With Sparse Grouping for 3D Point Cloud Semantic Segmentation in Mobile Devices von Kim, Sangjin, Kim, Sangyeob, Lee, Juhyoung, Yoo, Hoi-Jun

Volltext

DynaPlasia: An eDRAM In-Memory Computing-Based Reconfigurable Spatial Accelerator With Triple-Mode Cell von Kim, Sangjin, Li, Zhiyong, Um, Soyeon, Jo, Wooyoung, Ha, Sangwoo, Lee, Juhyoung, Kim, Sangyeob, Han, Donghyeon, Yoo, Hoi-Jun

Volltext

SRNPU: An Energy-Efficient CNN-Based Super-Resolution Processor With Tile-Based Selective Super-Resolution in Mobile Devices von Lee, Juhyoung, Lee, Jinsu, Yoo, Hoi-Jun

Volltext

DNPU: An Energy-Efficient Deep-Learning Processor with Heterogeneous Multi-Core Architecture von Shin, Dongjoo, Lee, Jinmook, Lee, Jinsu, Lee, Juhyoung, Yoo, Hoi-Jun

Volltext

TSUNAMI: Triple Sparsity-Aware Ultra Energy-Efficient Neural Network Training Accelerator With Multi-Modal Iterative Pruning von Kim, Sangyeob, Lee, Juhyoung, Kang, Sanghoon, Han, Donghyeon, Jo, Wooyoung, Yoo, Hoi-Jun

Volltext

A Power-Efficient CNN Accelerator With Similar Feature Skipping for Face Recognition in Mobile Devices von Kim, Sangyeob, Lee, Juhyoung, Kang, Sanghoon, Lee, Jinsu, Yoo, Hoi-Jun

Volltext

ECIM: Exponent Computing in Memory for an Energy-Efficient Heterogeneous Floating-Point DNN Training Processor von Lee, Juhyoung, Kim, Jihoon, Jo, Wooyoung, Kim, Sangyeob, Kim, Sangjin, Yoo, Hoi-Jun

Volltext

OmniDRL: An Energy-Efficient Deep Reinforcement Learning Processor With Dual-Mode Weight Compression and Sparse Weight Transposer von Lee, Juhyoung, Kim, Sangyeob, Kim, Sangjin, Jo, Wooyoung, Kim, Ji-Hoon, Han, Donghyeon, Yoo, Hoi-Jun

Volltext

이미지 데이터를 모니터링하는 관리도에서 이미지와 ROI 크기 조정의 영향 von 이주형, Juhyoung Lee, 윤형욱, Hyeonguk Yoon, 이성민, Sungmin Lee, 이재헌, Jaeheon Lee

Volltext

A Mobile DNN Training Processor With Automatic Bit Precision Search and Fine-Grained Sparsity Exploitation von Han, Donghyeon, Im, Dongseok, Park, Gwangtae, Kim, Youngwoo, Song, Seokchan, Lee, Juhyoung, Yoo, Hoi-Jun

Volltext

PNPU: An Energy-Efficient Deep-Neural-Network Learning Processor With Stochastic Coarse-Fine Level Weight Pruning and Adaptive Input/Output/Weight Zero Skipping von Kim, Sangyeob, Lee, Juhyoung, Kang, Sanghoon, Lee, Jinmook, Jo, Wooyoung, Yoo, Hoi-Jun

Volltext

An Energy-Efficient Sparse Deep-Neural-Network Learning Accelerator With Fine-Grained Mixed Precision of FP8-FP16 von Lee, Jinsu, Lee, Juhyoung, Han, Donghyeon, Lee, Jinmook, Park, Gwangtae, Yoo, Hoi-Jun

Volltext