Yonsei University

About: Yonsei University is a education organization based out in Seoul, South Korea. It is known for research contribution in the topics: Population & Cancer. The organization has 50162 authors who have published 106172 publications receiving 2279044 citations. The organization is also known as: Yonsei.

Learning Memory-Guided Normality for Anomaly Detection

Abstract: We address the problem of anomaly detection, that is, detecting anomalous events in a video sequence. Anomaly detection methods based on convolutional neural networks (CNNs) typically leverage proxy tasks, such as reconstructing input video frames, to learn models describing normality without seeing anomalous samples at training time, and quantify the extent of abnormalities using the reconstruction error at test time. The main drawbacks of these approaches are that they do not consider the diversity of normal patterns explicitly, and the powerful representation capacity of CNNs allows to reconstruct abnormal video frames. To address this problem, we present an unsupervised learning approach to anomaly detection that considers the diversity of normal patterns explicitly, while lessening the representation capacity of CNNs. To this end, we propose to use a memory module with a new update scheme where items in the memory record prototypical patterns of normal data. We also present novel feature compactness and separateness losses to train the memory, boosting the discriminative power of both memory items and deeply learned features from normal data. Experimental results on standard benchmarks demonstrate the effectiveness and efficiency of our approach, which outperforms the state of the art.

Cognitive Radio Networks with Energy Harvesting

Abstract: We consider a cognitive radio network with an energy-harvesting secondary transmitter to improve both energy efficiency and spectral efficiency. The goal of this paper is to determine an optimal spectrum sensing policy that maximizes the expected total throughput subject to an energy causality constraint and a collision constraint. The energy causality constraint comes from the fact that the total consumed energy should be equal to or less than the total harvested energy, while the collision constraint is required to protect the primary user. We first show that the system can be divided into a spectrum-limited regime and an energy-limited regime depending on where the detection threshold for the spectrum sensor lies. Assuming infinite battery capacity, we derive the optimal detection threshold that maximizes the expected total throughput subject to the energy causality constraint and the collision constraint. Analytical and numerical results show that the system is energy-limited if the energy arrival rate is lower than the expected energy consumption for a single spectrum access. They also show that a decreasing probability of accessing the occupied spectrum does not always result in decreased probability of accessing the idle spectrum in the energy-limited regime.

Adrenal peripheral clock controls the autonomous circadian rhythm of glucocorticoid by causing rhythmic steroid production

Abstract: Glucocorticoid (GC) is an adrenal steroid with diverse physiological effects. It undergoes a robust daily oscillation, which has been thought to be driven by the master circadian clock in the suprachiasmatic nucleus of the hypothalamus via the hypothalamus–pituitary–adrenal axis. However, we show that the adrenal gland has its own clock and that the peripheral clockwork is tightly linked to steroidogenesis by the steroidogenic acute regulatory protein. Examination of mice with adrenal-specific knockdown of the canonical clock protein BMAL1 reveals that the adrenal clock machinery is required for circadian GC production. Furthermore, behavioral rhythmicity is drastically affected in these animals, together with altered expression of Period1, but not Period2, in several peripheral organs. We conclude that the adrenal peripheral clock plays an essential role in harmonizing the mammalian circadian timing system by generating a robust circadian GC rhythm.

The Y^2 Stellar Evolutionary Tracks

Abstract: We present a database of the latest stellar models of the $Y^2$ (Yonsei-Yale) collaboration. This database contains the stellar evolutionary tracks from the pre-main-sequence birthline to the helium core flash that were used to construct the $Y^2$ isochrones. We also provide a simple interpolation routine that generates stellar tracks for given sets of parameters (metallicity, mass, and $\alpha$-enhancement).

Incorporation of Pazopanib in Maintenance Therapy of Ovarian Cancer

Abstract: Purpose Pazopanib is an oral, multikinase inhibitor of vascular endothelial growth factor receptor (VEGFR) -1/-2/-3, platelet-derived growth factor receptor (PDGFR) -α/-β, and c-Kit. Preclinical and clinical studies support VEGFR and PDGFR as targets for advanced ovarian cancer treatment. This study evaluated the role of pazopanib maintenance therapy in patients with ovarian cancer whose disease did not progress during first-line chemotherapy. Patients and Methods Nine hundred forty patients with histologically confirmed cancer of the ovary, fallopian tube, or peritoneum, International Federation Gynecology Obstetrics (FIGO) stages II-IV, no evidence of progression after primary therapy consisting of surgery and at least five cycles of platinum-taxane chemotherapy were randomized 1:1 to receive pazopanib 800 mg once per day or placebo for up to 24 months. The primary end point was progression-free survival by RECIST 1.0 assessed by the investigators. Results Maintenance pazopanib prolonged progression-fre...