Jiwon Choi

Jiwon Choi

Auto-generated profile, based on publicly available information

Jiwon Choi

School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 November 2023 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
SKSangyeob Kim
Sangyeob Kim
SKSoyeon Kim
Soyeon Kim
SHSeongyon Hong
Seongyon Hong
SKSangjin Kim
Sangjin Kim
DHDonghyeon Han
Donghyeon Han
HYHoi-Jun Yoo
Hoi-Jun Yoo
In this article, we propose a complementary deep-neural-network (C-DNN) processor by combining convolutional neural network (CNN) and spiking neural network (SNN) to take advantage of them. The C-DNN processor can support both complementary inference and training with heterogeneous CNN and SNN core architecture. In addition, the C-DNN processor is the first DNN accelerator application-specific integrated circuit (ASIC) that can support CNN–SNN workload division by using their magnitude–energy tradeoff. The C-DNN processor integrates the CNN–SNN workload allocator and attention module to find a more energy-efficient network domain for each workload in DNN. They enable the C-DNN processor to operate at the energy optimal point. Moreover, the SNN processing element (PE) array with distributed L1 cache can reduce the redundant memory access for SNN processing, resulting in a 42.2%–49.1% reduction. For high energy-efficient DNN training, the C-DNN processor integrates the global counter and local delta-weight (LDW) unit to eliminate power-consuming counters for a forward delta-weight generation. Furthermore, the forward delta-weight-based sparsity generation (FDWSG) is proposed to reduce the number of operations for training by 31%–79% The C-DNN processor achieves an energy efficiency of 85.8 and 79.9 TOPS/W for inference with CIFAR-10 and CIFAR-100, respectively (VGG-16). Moreover, the C-DNN processor achieves ImageNet classification with state-of-the-art energy efficiency of 24.5 TOPS/W (ResNet-50). For training, the C-DNN processor achieves the state-of-the-art energy efficiency of 84.5 and 17.2 TOPS/W for CIFAR-10 and ImageNet, respectively. Furthermore, it achieves 77.1% accuracy for ImageNet training with ResNet-50.

25 March 2019 in Antonie van Leeuwenhoek
Verified
Journal Article
Closed access
PDF file
IKInhyup Kim
Inhyup Kim
JCJiwon Choi
Jiwon Choi
GCGeeta Chhetri
Geeta Chhetri
Taegun SeoTaegun Seo
Taegun Seo

19 April 2022 in Journal of Lightwave Technology
Verified
Journal Article
Closed access
GSGyeongho Son
Gyeongho Son
RPRizki Arif Pradono
Rizki Arif Pradono
JCJiwon Choi
Jiwon Choi
YJYoungjae Jeong
Youngjae Jeong
DHDae Seok Han
Dae Seok Han
KYKyoungsik Yu
Kyoungsik Yu
The adiabatically tapered subwavelength-scale optical fiber tapers (OFTs) have been of interest for many photonic applications. We demonstrate chemical etching-based adiabatic OFTs and their relaxed adiabaticity conditions over broad operation wavelength ranges. The etched OFTs show adiabatically-varying tapering angles as small as 1 and mechanical flexibility. When compared to the conventional methods, the proposed OFTs and their fabrication techniques may allow reproducible batch production routes for adiabatic OFTs with enhanced uniformity, enabling a variety of applications in subwavelength-scale light-matter interactions and efficient light coupling between waveguides.

22 September 2025 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Verified
Journal Article
Closed access
JOJungjun Oh
Jungjun Oh
SKSangjin Kim
Sangjin Kim
JCJiwon Choi
Jiwon Choi
JRJunha Ryu
Junha Ryu
BKByeongcheol Kim
Byeongcheol Kim
HYHoi-Jun Yoo
Hoi-Jun Yoo
Token merging is a widely adopted technique in vision transformers (ViTs) to reduce computation with minimal accuracy degradation. However, in tasks that require fine-grained outputs, such as text-unified object recognition, aggressive merging reaches a limit due to accuracy loss. To address this, we introduce a method that exploits residual token similarity beyond the merging threshold using difference computing, which significantly improves energy efficiency. The proposed processor integrates the post merging residual difference (PMRD) algorithm, which first reduces 34% of tokens through merging, then applies difference computing to the remaining tokens. This results in a 31% reduction in effective bit operations by leveraging residual similarity. To further exploit the resulting bit-level sparsity, the processor employs sparsity-separated accumulation with sign-magnitude representation (SSA-SM) and a bi-directional dynamic memory allocation (BDA) to optimize both computation and memory access. These enhancements yield an additional 29.4% computational energy savings. Overall, the system achieves a $2.22\times $ improvement in energy efficiency and delivers 7.23 TOPS/W on the MS COCO dataset for text-unified object recognition.

6 November 2023 in IEEE Micro
Verified
Journal Article
Closed access
SKSangyeob Kim
Sangyeob Kim
SKSoyeon Kim
Soyeon Kim
SHSeongyon Hong
Seongyon Hong
SKSangjin Kim
Sangjin Kim
JCJiwon Choi
Jiwon Choi
HYHoi-Jun Yoo
Hoi-Jun Yoo
The authors propose a complementary online learning neural processing unit (COOL-NPU) to implement a highly accurate and high-energy-efficient online learning system. It reduces the energy consumption by combining the training methods of CNN and SNN, and eliminates the power overhead due to the redundant weight update by training trigger with SNN-gradient. The proposed SNN core reduces the energy consumption of SNN-gradient generation by 2-step encoding, and reduces inference power by hierarchical cache with LUT-mode. In addition, it supports neuron-level event-driven backward operation to maximize the effect of the training trigger. Fabricated with Samsung 28-nm CMOS technology, the COOL-NPU achieves 6.94 mJ/frame and 0.73 mAP for object detection, resulting in 47.7% energy reduction with a slight accuracy loss compared to previous state-of-the-art.

22 May 2026 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
JCJiwon Choi
Jiwon Choi
SHSeongyon Hong
Seongyon Hong
WJWooyoung Jo
Wooyoung Jo
WPWonhoon Park
Wonhoon Park
SWSunjoo Whang
Sunjoo Whang
HYHoi-Jun Yoo
Hoi-Jun Yoo
Digital compute-in-memory (DCIM) and the microscaling (MX) data format have emerged as promising hardware and algorithmic solutions, respectively, for energy-efficient generative AI (GenAI) processing. However, their combination introduces challenges, including redundant computation from mantissa alignment, excessive exponent handling overhead, and accuracy loss from static approximation schemes that fail to accommodate block-wise dynamic bit-significance. In this article, we present MIDAS, an MX-optimized DCIM-based GenAI accelerator with three key features: 1) spatio-temporal cyclic alignment (STCA) with sign-magnitude-based assembled column addition (SACA) enables compact weight mapping, high input reusability, and low-latency computation, achieving $1.72\times $ higher macro energy efficiency; 2) MX-aware minimum exponent post-detection (MEPD) eliminates redundant maximum exponent detection by exploiting the inherent characteristics of MX quantization, reducing exponent handling power, area, and latency by 68.8%, 63.2%, and 22.5%, respectively; and 3) sense amplifier-based dynamic bit (SADB)-significance handling with shared scale-aware channel regrouping (SSCR) dynamically adjusts approximation bit-positions according to each block’s shared scale, improving energy efficiency by 27.3% while minimizing accuracy loss. MIDAS is fabricated in Samsung 28-nm CMOS technology and occupies a 7.56 mm2 die area. MIDAS achieves macro energy and area efficiencies of 96.3 TFLOPS/W and 0.69 TFLOPS/mm2, respectively, under MXFP4. Compared to state-of-the-art DCIMs, MIDAS demonstrates $1.58\times $ higher system energy efficiency and $3.32\times $ higher macro energy efficiency, validating its effectiveness as an MX-optimized AI accelerator for next-generation GenAI applications.