이수민 교수

Predicting both accurate and reliable solubility values has long been a crucial but challenging task. In this work, surrogated model-based methods were developed to accurately predict the solubility of two molecules (solute and solvent) through machine learning and deep learning. The current study employed two methods: (1) converting molecules into molecular fingerprints and adding optimal physicochemical properties as descriptors and (2) using graph convolutional network (GCN) models to convert

While deep learning (DL) has brought a revolution in the protein structure prediction field, still an important question remains how the revolution can be transferred to advances in structure-based drug discovery. Because the lessons from the recent GPCR dock challenge were inconclusive primarily due to the size of the dataset, in this work we further elaborated on 70 diverse GPCR complexes bound to either small molecules or peptides to investigate the best-practice modeling and docking strategi

Technologies to decipher cellular biology, such as bulk sequencing technologies and single-cell sequencing technologies, have greatly assisted novel findings in tumor biology. Recent findings in tumor biology suggest that tumors construct architectures that influence the underlying cancerous mechanisms. Increasing research has reported novel techniques to map the tissue in a spatial context or targeted sampling-based characterization and has introduced such technologies to solve oncology regardi

Detecting fashion landmarks is a fundamental technique for visual clothing analysis. Due to the large variation and non-rigid deformation of clothes, localizing fashion landmarks suffers from large spatial variances across poses, scales, and styles. Therefore, understanding contextual knowledge of clothes is required for accurate landmark detection. To that end, in this paper, we propose a fashion landmark detection network with a global-local embedding module. The global-local embedding module

Pen drawing is a method that allows simple, inexpensive, and intuitive two-dimensional (2D) fabrication. To integrate such advantages of pen drawing in fabricating 3D objects, we developed a 3D fabrication technology that can directly transform pen-drawn 2D precursors into 3D geometries. 2D-to-3D transformation of pen drawings is facilitated by surface tension-driven capillary peeling and floating of dried ink film when the drawing is dipped into an aqueous monomer solution. Selective control of

The convolutional neural network (CNN)-based super-resolution (SR) has shown outstanding performance in the field of computer vision. The implementation of inference hardware for CNN-based SR has suffered from the intensive computation with severely unbalanced computation load among layers. Various light-weighted SR networks have been researched with little performance degradation. However, the hardware-efficient dataflow is also required to efficiently accelerate inference hardware within limit

In multi-modal action recognition, it is important to consider not only the complementary nature of different modalities but also global action content. In this paper, we propose a novel network, named Modality Mixer (M-Mixer) network, to leverage complementary information across modalities and temporal context of an action for multi-modal action recognition. We also introduce a simple yet effective recurrent unit, called Multi-modal Contextualization Unit (MCU), which is a core component of M-M

Sumin Lee, Jae-Won Choi, Kyoung-Min Kim, Jun Won Kim, Sooyeon Kim, Taewoong Kang, Johanna Inhyang Kim, Young Sik Lee, Bongseog Kim, Doug Hyun Han, Jae Hoon Cheong, Soyoung Irene Lee, Gi Jung Hyun, and Bung-Nyun Kim. J Korean Acad Child Adolesc Psychiatry 2016;27:236-66. https://doi.org/10.5765/jkacap.2016.27.4.236

This paper proposes a convolutional neural network (CNN)-based super-resolution accelerator for up-scaling to ultra-HD (UHD) resolution in real-time in edge devices. A novel error-compensated bit quantization is adopted to reduce bit depth in the SR task. Spatially independent layer fusion is exploited to satisfy high throughput requirements at UHD resolution by increasing parallelism. Burst operation with write mask in the dual-port SRAM increases the process element utilization by allowing the

Carbon corrosion in a catalyst layer (CL) deteriorates the performance and durability of proton‐exchange membrane fuel cells (PEMFCs), which are closely related to water management within these cells. This study investigates the characteristics of water behavior of two gas diffusion layers (GDLs) and compares their influence on degrees of degradation in the CL. First, the properties of the GDLs, including their thickness, pore size distribution, gas permeability, electrical resistance, contact a

A series elastic actuator (SEA) is mainly used in human-robot interaction applications. Especially, a reaction-force-sensing SEA (RFSEA), where the spring is located between the ground and the actuator, has been developed as a practical implementation of the SEAs, as their form-factors are superior to the conventional SEAs. However, the RFSEA has limitations on its torque control performance. The output torque of an RFSEA is estimated as the spring torque, assuming the load is fixed. However, si

Online action detection, which aims to identify an ongoing action from a streaming video, is an important subject in real-world applications. For this task, previous methods use recurrent neural networks for modeling temporal relations in an input sequence. However, these methods overlook the fact that the input image sequence includes not only the action of interest but background and irrelevant actions. This would induce recurrent units to accumulate unnecessary information for encoding featur

Encoded microparticles have great potential in small-volume multiplexed assays. It is important to link the micro-level assays to the macro-level by indexing and manipulating the microparticles to enhance their versatility. There are technologies to actively manipulate the encoded microparticles, but none is capable of directly manipulating the encoded microparticles with homogeneous physical properties. Here, we report the image-based laser-induced forward transfer system for active manipulatio