이종혁 교수

In this paper, we present a two-stage neural quality estimation model that uses multilevel task learning for translation quality estimation (QE) at the sentence, word, and phrase levels. Our approach is based on an end-to-end stacked neural model named Predictor-Estimator, which has two stages consisting of a neural word prediction model and neural QE model. To efficiently train the two-stage model, a stack propagation method is applied, thereby enabling us to jointly learn the word prediction m

This paper presents a novel approach using recurrent neural networks for estimating the quality of machine translation output. A sequence of vectors made by the prediction method is used as the input of the final recurrent neural network. The prediction method uses bi-directional recurrent neural network architecture both on source and target sentence to fully utilize the bi-directional quality information from source and target sentence. Our experiments show that the proposed recurrent neural n

Side-channel analysis is a serious type of attack that can break mathematically secure cryptographic algorithms. Many studies have designed countermeasures against side-channel analysis, such as masking and hiding schemes. Frequently, designers employ combined countermeasures that use both a first-order masking scheme and a hiding scheme to provide sufficient security and efficiency. Random insertion of dummy operations scheme, which is one of the hiding schemes, randomly changes the execution t

Fine segmentation labelling tasks are time consuming and typically require a great deal of manual labor. This paper presents a novel method for efficiently creating pixel-level fine segmentation labelling that significantly reduces the amount of necessary human labor. The proposed method utilizes easily produced multiple and complementary coarse labels to build a complete fine label via supervised learning. The primary label among the coarse labels is the manual label, which is produced with sim

We studied the effect of structural and morphological changes on the conductivity of a stretched conducting polymer film. To improve the poor processability of polyaniline, we used dodecylbenzenesulfonic acid as both a surfactant and a dopant during emulsion polymerization, followed by blending with high-impact polystyrene. UV-Vis/NIR spectra were obtained to observe conformational changes, and SEM and AFM were used to investigate morphological changes. FT-IR dichroism was applied to determine t

In this paper, we propose a fast phase synchronization method with clustering and one bit feedback for distributed beamforming in a wireless sensor network where multiple single-antenna nodes transfer their sensing data to a data collector. To transmit the data effectively, the distributed nodes should adjust their transmit signals' phase properly so that their signals are constructively added up at the data collector with a same phase, resulting in an increasing received signal strength (RSS).

To estimate angle, velocity, and range information of multiple targets jointly in FMCW MIMO radar, two-dimensional (2D) MUSIC with matched filtering and FFT algorithm is proposed. By reformulating the received FMCW signal of the colocated MIMO radar, we exploit 2D MUSIC to estimate the angle and Doppler frequency of multiple targets. Then by using a matched filter together with the estimated angle and Doppler frequency and FFT operation, the range of the target is estimated. To effectively estim