Machine Intelligence and Information Learning (MIIL) lab’s paper titled “XtarNet: Learning to Extract Task-Adaptive Representation for Incremental Few-Shot Learning” is accepted to International Conference on Machine Learning (ICML), which is one of top conferences for artificial intelligence (acceptance rate 21.8%). Learning novel concepts while preserving prior knowledge is a long-standing challenge in machine learning. The challenge gets

One BMIPL’s paper, “Multi-Temporal Recurrent Neural Networks For Progressive Non-Uniform Single Image Deblurring With Incremental Temporal Training,” by Dongwon Park, Dong Un Kang, Jisoo Kim and Se Young Chun has been accepted to European Conference on Computer Vision (ECCV) 2020 (acceptance rate: 27%) as a spotlight presentation (top 5% of submissions). Spotlight presentation is given

One BMIPL’s paper, “Unsupervised Training Of Deep Low-Dose CT Reconstruction Without Full-Dose CT Images,” by Kwanyoung Kim, Shakarim Soltanayev and Se Young Chun has been accepted to IEEE Journal of Selected Topics In Signal Processing (IF 6.688). Kwanyoung Kim is about to finish his MS program at BMIPL, UNIST, and Shakarim Soltanayev finished his MS

Professor Woongki Baek in the department of Computer Science and Engineering at UNIST has been selected as a recipient of Most Influential PLDI Paper Award at PLDI 2020 for his research paper entitled “Green: A Framework for Supporting Energy-Conscious Programming using Controlled Approximation”, which was published in PLDI 2010. PLDI (ACM SIGPLAN Conference on Programming

A paper have been accepted for publication at ATC 2020. This paper is co-authored by Jay H. Park, Gyeongchan Yun, Chang M. Yi, Nguyen T. Nguyen, Seungmin Lee, Jaesik Choi (KAIST), Sam H. Noh, and Young-ri Choi.   A research paper entitled “HetPipe: Enabling Large DNN Training on (Whimpy) Heterogeneous GPU Clusters through Integration of

Prof. Seung Ryul Baek has joined the School of ECE (CSE Track) as of Apr.01, 2020. I am truly happy to join UNIST as the faculty member of the school of Electrical and Computer Engineering. My research interest generally lies in the field of computer vision and deep learning, especially on the application of 3D

Prof. Sung Whan Yooneong hwan has joined the School of ECE (EE Track) as of Mar.01, 2020. I am truly delighted to join School of Electrical and Computer Engineering of UNIST as a faculty member. My research interests are in the area of artificial intelligence, information/communication theory and distributed systems, with emphasis on developing learning algorithms

Journal title: IEEE Transactions on Antennas and Propagations Condition: Accepted Article title: Geometry-Independent Excitation of Dark Modes Using Dipole Moment Transitions Authors: Thi Hai-Yen Nguyen, Hoon Yeub Jeong, Young Chul Jun, and Gangil Byun Abstract This article proposes a novel approach to adaptive transitions between different dipole moments to allow for consistent excitation of Fano

UNIST ECE student, Kyu-Yul Lee and Prof. Jae-Young Sim’s paper has been accepted to CVPR 2020.   Authors: Kyu-Yul Lee and Jae-Young Sim (corresponding author) Title: Warping Residual Based Image Stitching for Large Parallax Conference: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)   Image stitching techniques align two images captured at different

Journal title: IEEE Transactions on Power Electronics (2-yr IF: 7.224, 5-yr IF: 8.446, 5.47%, L1) Condition: Accepted Article title: Practical Design Methodology of IH and IPT Dual-Functional Apparatus Authors: Mina Kim, Hwa-Pyeong Park, and Jee-Hoon Jung* Abstract A domestic induction heating (IH) apparatus uses series resonance to transfer the electric energy to the IH vessel. Similarly,

Journal title: IEEE Transactions on Industrial Electronics (2-yr IF: 7.050, 5-yr IF: 7.623, 5.00%, L1) Condition: Accepted Article title: Effective Magnetic Component Design of Three Phase Dual Active Bridge Converter for LVDC Distribution System Authors: Hyun-Jun Choi, Bong-Gyo Seo, Myung-Hyo Ryu,  Young-Pyo Cho, and Jee-Hoon Jung* Abstract To improve the power conversion efficiency of a three-phase dual active

A research paper entitled “Hotness- and Lifetime-Aware Data Placement and Migration for High-Performance Deep Learning on Heterogeneous Memory Systems” has been accepted for publication in IEEE Transactions on Computers. The paper is co-authored by Myeonggyun Han, Jihoon Hyun, Seongbeom Park, and Prof. Woongki Baek at Computer Architecture and Systems Lab. (CASL), CSE, UNIST. This work

UNIST ECE student Ramesh Patel, research fellow Dr. Min Woo Ryu and Prof. Kyung Rok Kim (with Prof. Ki Jin Han, Dongguk Univ.)’s paper has been accepted to IEEE Transactions on Antennas and Propagation (IF: 4.435). IEEE Transactions on Antennas and Propagation (TAP) is the most prestigious journal on the state-of-the-art works in the field

One BMIPL’s paper, “Algorithms and Analyses for Joint Spectral Image Reconstruction in Y-90 Bremsstrahlung SPECT,” by Se Young Chun, Minh Phuong Nguyen, Thanh Quoc Phan, Hanvit Kim, Jeffrey A Fessler and Yuni K. Dewaraja, has been accepted to IEEE Transactions on Medical Imaging (IF: 7.816). This is a collaboration work between UNIST and University of Michigan

UNIST ECE student Seoktae Seo, Hyunkyeong Jo and Prof. Franklin Bien’s paper has been accepted to IEEE Transactions on Power Electronics A free arrangement wireless power transfer (WPT) system with a ferrite transmission medium and geometry-based performance improvement method are introduced here. The proposed system addresses the efficiency collapse caused by misalignment or increased distance

UNIST ECE student Jae-Seong Yun and Prof. Jae-Young Sim’s paper has been accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (IF: 17.730). Large-scale 3D point clouds (LS3DPCs) captured by terrestrial LiDAR (Light Detection And Ranging) scanners often include virtual points which are generated by glass reflection. The virtual points may degrade the performance

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

Top