One BMIPL’s paper, “Extending Stein’s unbiased risk estimator to train deep denoisers with correlated pairs of noisy images,” by Magauiya Zhussip, Shakarim Soltanayev and Se Young Chun has been accepted to NeurIPS 2019 (acceptance rate: 21.2%). Magauiya Zhussip and Shakarim Soltanayev just finished their MS program at BMIPL, UNIST. NeurIPS is one of the top machine learning conferences.

A research paper entitled “MOSAIC: Heterogeneity-, Communication-, and Constraint-Aware Model Slicing and Execution for Accurate and Efficient Inference” has been accepted for publication at PACT’19. The paper is co-authored by Myeonggyun Han, Jihoon Hyun, Seongbeom Park, Jinsu Park, and Prof. Woongki Baek at Computer Architecture and Systems Lab. (CASL), CSE, UNIST. This work proposes MOSAIC,

A poster of BMIPL, UNIST and University of Michigan won the 2nd Place Poster Award on Physics, Instrumentation and Data Sciences at the 2019 annual meeting of Society of Nuclear Medicine and Molecular Imaging. A collaboration paper titled “On Enhancing Monte‐Carlo Scatter Correction for Y90 Bremsstrahlung SPECT using Guided Filtering” by Hanvit Kim (the first author, PhD student at

Students of BMIPL (Bio-Medical Image Processing Lab.) won five awards at NTIRE 2019 Challenges ( of CVPR (Computer Vision and Pattern Recognition, one of the top conferences in computer vision) under the supervision of Prof. Se Young Chun. Dongwon Park (MS program) won a Winner Award for NTIRE 2019 Challenge on Real Image Super-Resolution, a Runner-Up Award for NTIRE 2019 Challenge on Image

The paper authored by ECE students of iVADER Lab.(Chunggi Lee, Yeonjun Kime, Dongmin Kim, advisor: Prof. Sungahn Ko), “A Visual Analytics System for Exploring, Monitoring, and Forecasting Road Traffic Congestion” has been accepted at IEEE Transaction on Visualization and Computer Graphics. This paper is invited for oral presentation at the forthcoming 2019 IEEE VIS, which

Authors: Jiyeon Han*, Kyowoon Lee* , Anh Tong, Jaesik Choi (corresponding author) (* contributed equally) Title: Confirmatory Bayesian Online Change Point Detection in the Covariance Structure of Gaussian Processes Conference: International Joint Conference on Artificial Intelligence (IJCAI 2019)  A research paper entitled “Confirmatory Bayesian Online Change Point Detection in the Covariance Structure of Gaussian Processes”

Prof. Jeon’s paper titled “Analysis of Large-Scale Multi-Tenant GPU Clusters for DNN Training Workloads” has been accepted to USENIX ATC 2019. This work was done in collaboration with researchers at Microsoft in US and China. This paper presents an in-depth analysis on a multi-tenant GPU cluster running deep learning workloads, with discussion focused on efficiency

– Authors : Anh Tong, Jaesik Choi (corresponding author) – Title: Discovering Latent Covariance Structures for Multiple Time Series – Conference :  International Conference on Machine Learning (ICML 2019) A research paper entitled “Discovering Latent Covariance Structures for Multiple Time Series” has been accepted to ICML(International Conference on Machine Learning) 2019. The paper is authored by Anh

A paper entitled “A Low Noise Single-Slope ADC with Signal-Dependent Multiple Sampling Technique” has been accepted to International Image Sensor Workshop (IISW) in this year. The paper is authored by Sanguk Lee, Seunghyun Lee, Bumjun Kim, and Prof. Seong-Jin Kim. IISW is one of the most prestigious conferences in the field of CMOS imagers and

Prof. Seong-Jin Kim has been invited to serve as a technical program committee (TPC) member at the IEEE International Solid-State Circuit Conference (ISSCC). As the single most prestigious conference in the solid-state circuit society, ISSCC is also called “the Olympic in the field of the semiconductor circuit,” and it is held in San Francisco every

Sam H. Noh, UNIST Professor of Computer Science and Engineering, has been elected to serve as one of two co-chairs for the USENIX ’18th USENIX Conference on File and Storage Technologies’ (FAST ’20). This can be viewed as a positive sign that South Korean data storage technologies, including flash memory are world-class academically. USENIX, founded in 1975, is the Advanced

A research paper entitled “CoPart: Coordinated Partitioning of Last-Level Cache and Memory Bandwidth for Fairness-Aware Workload Consolidation on Commodity Servers” has been accepted for publication at EuroSys’19. The paper is co-authored by Jinsu Park, Seongbeom Park, and Prof. Woongki Baek at Computer Architecture and Systems Lab. (CASL), CSE, UNIST. This work proposes CoPart, coordinated partitioning

Two papers have been accepted for publication at FAST 2019. One paper is co-authored by Olzhas Kaiyrakhmet, Songyi Lee, Prof. Beomseok Nam (Sungkyunkwan Univ.), Prof. Sam H. Noh, and Prof. Young-ri Choi, and the other paper is co-authored by Moohyeon Nam, Hokeun Cha (Sungkyunkwan Univ.), Prof. Young-ri Choi,  Prof. Sam H. Noh, and Prof. Beomseok

  Abstract In this talk, I will introduce both completed work and work in progress in the intersection of systems and machine learning. After (very) briefly discussing projects on reducing tail latency in interactive services and intelligent real-time data analytics, I will dive into the main body of the talk, which is toward building GPU

Abstract Despite the advances in understandings and mechanisms for securing computer systems, researchers are reporting new forms of attacks every year, and production systems often do not adopt all the state-of-the-art techniques. The fundamental reason behind the insecurity is the ever-increasing complexity of the systems, which is unavoidable because of programmability and time to market.

Professor Myeongjae Jeon’s paper titled “Tiresias: A GPU Cluster Manager for Distributed Deep Learning” has been accepted to NSDI 2019. This work was done in collaboration with researchers at University of Michigan, Microsoft Research, Alibaba, and Bytedance.   This paper presents Tiresias, a GPU cluster resource manager tailored for distributed deep learning training, which efficiently

Professor Hyungon Moon’s paper titled “Fuzzing File Systems via Two-Dimensional Input Space Exploration” has been accepted to IEEE S&P (Oakland) 2019. This work was done while he was a postdoctoral fellow at Georgia Tech. This paper presents Janus, a fuzzer that mutates both the file operations and the file system images to find bugs in

41 NRF projects selected as excellent outcomes of the year by the Ministry of Education. Out of the 14 awardees in the field of science and engineering, UNIST takes 2 wins. ————— The Ministry of Education and NRF held a joint award ceremony on November 13th in Seoul announcing the list of 41 projects selected

The first authors: Juyeop Kim, Heein Yoon, Younghyun Lim, and Seyeon Yoo in the Ph.D. program of ECE (From the left side)   Two ICSL (Prof. Jaehyouk Choi)’s papers have been accepted for presentation at IEEE ISSCC (International Solid-State Circuits Conference) in San Francisco in February 2019.   ISSCC is the single most prestigious conference

연구논문 어떻게 써야할지 모르겠다면? 저널 리젝(reject)에 어떻게 대처해야할지 모르겠다면? 지금 바로 연구논문작성 특강에 참가해보세요! 여러분의 많은 관심과 참여 부탁드립니다.   The library will be holding a special lecture entitled “How to Write a Research Paper.” You will be introduced to academic publishing and how to deal with rejection from journals. If you’ve never written a