• Title, Summary, Keyword: semi-convergence

Search Result 221, Processing Time 0.033 seconds

Mobile User Interface Pattern Clustering Using Improved Semi-Supervised Kernel Fuzzy Clustering Method

  • Jia, Wei;Hua, Qingyi;Zhang, Minjun;Chen, Rui;Ji, Xiang;Wang, Bo
    • Journal of Information Processing Systems
    • /
    • v.15 no.4
    • /
    • pp.986-1016
    • /
    • 2019
  • Mobile user interface pattern (MUIP) is a kind of structured representation of interaction design knowledge. Several studies have suggested that MUIPs are a proven solution for recurring mobile interface design problems. To facilitate MUIP selection, an effective clustering method is required to discover hidden knowledge of pattern data set. In this paper, we employ the semi-supervised kernel fuzzy c-means clustering (SSKFCM) method to cluster MUIP data. In order to improve the performance of clustering, clustering parameters are optimized by utilizing the global optimization capability of particle swarm optimization (PSO) algorithm. Since the PSO algorithm is easily trapped in local optima, a novel PSO algorithm is presented in this paper. It combines an improved intuitionistic fuzzy entropy measure and a new population search strategy to enhance the population search capability and accelerate the convergence speed. Experimental results show the effectiveness and superiority of the proposed clustering method.

EER-ASSL: Combining Rollback Learning and Deep Learning for Rapid Adaptive Object Detection

  • Ahmed, Minhaz Uddin;Kim, Yeong Hyeon;Rhee, Phill Kyu
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • v.14 no.12
    • /
    • pp.4776-4794
    • /
    • 2020
  • We propose a rapid adaptive learning framework for streaming object detection, called EER-ASSL. The method combines the expected error reduction (EER) dependent rollback learning and the active semi-supervised learning (ASSL) for a rapid adaptive CNN detector. Most CNN object detectors are built on the assumption of static data distribution. However, images are often noisy and biased, and the data distribution is imbalanced in a real world environment. The proposed method consists of collaborative sampling and EER-ASSL. The EER-ASSL utilizes the active learning (AL) and rollback based semi-supervised learning (SSL). The AL allows us to select more informative and representative samples measuring uncertainty and diversity. The SSL divides the selected streaming image samples into the bins and each bin repeatedly transfers the discriminative knowledge of the EER and CNN models to the next bin until convergence and incorporation with the EER rollback learning algorithm is achieved. The EER models provide a rapid short-term myopic adaptation and the CNN models an incremental long-term performance improvement. EER-ASSL can overcome noisy and biased labels in varying data distribution. Extensive experiments shows that EER-ASSL obtained 70.9 mAP compared to state-of-the-art technology such as Faster RCNN, SSD300, and YOLOv2.

An analysis and modification of a unified phase 1-phase 2 semi-infinite constrained optimization algorithm

  • Yang, Hyun-Suk
    • 제어로봇시스템학회:학술대회논문집
    • /
    • /
    • pp.483-487
    • /
    • 1994
  • In this paper, we analize the effect of a steering water used in a unified phase I-phase II semi-infinite constrained optimization algorithm and present a new algorithm based on the facts that when the point x is far away from the feasible region where all the constraints are satisfied, reaching to the feasible region is more important than minimizing the cost function and that when the point x is near the region, it is more efficient to try to reach the feasible region and to minimize the cost function concurrently. Also, the angle between the search direction vector and the gradient of the cost function is considered when the steering parameter value is computed. Even though changing the steering parameter does not change the rate of convergence of the algorithm, we show through some examples that the proposed algorithm performs better than the other algorithms.

  • PDF

A Unified Phase I - Phase II Semi-Infinite Constrained Optimization Algorithm with a Varying Steering Parameter (가변의 조정변수를 갖는 복합된 1-2 단계 최적화 알고리즘)

  • Yang, Hyun-Suk
    • Journal of the Korean Institute of Telematics and Electronics B
    • /
    • v.31B no.9
    • /
    • pp.27-35
    • /
    • 1994
  • It is known that a unified phase I-phase II semi-infinite optimization algorithm with a steerign parameter performs better than the original unified phase I-phase II algorithm. In this paper, the effect of the steering parameter is analized and a new algorithm is presented based on the facts that when the point x is far away from the feasible region, reaching to the feasible region is more important than minimizing the cost functio and that when the point x is near the region, it is more efficient to try to reach the feasible region and to minimize the cost function concurrently. It is also important to consider the relationship between the feasible direction and the gradient of the cost function. Even though changing the steering parameter does not change the rate of convergence of the algorithm, it is shown from examples that given new algorithm is more efficient than the previous ones.

  • PDF

IMPLEMENTATION OF A SECOND-ORDER INTERPOLATION SCHEME FOR THE CONVECTIVE TERMS OF A SEMI-IMPLICIT TWO-PHASE FLOW ANALYSIS SOLVER (물-기체 2상 유동 해석을 위한 Semi-Implicit 방법의 대류항에 대한 이차정확도 확장)

  • Cho, H.K.;Lee, H.D.;Park, I.K.;Jeong, J.J.
    • 한국전산유체공학회:학술대회논문집
    • /
    • /
    • pp.290-297
    • /
    • 2009
  • A two-phase (gas and liquid) flow analysis solver, named CUPID, has been developed for a realistic simulation of transient two-phase flows in light water nuclear reactor components. In the CUPID solver, a two-fluid three-field model is adopted and the governing equations are solved on unstructured grids for flow analyses in complicated geometries. For the numerical solution scheme, the semi-implicit method of the RELAP5 code, which has been proved to be very stable and accurate for most practical applications of nuclear thermal hydraulics, was used with some modifications for an application to unstructured non-staggered grids. This paper is concerned with the effects of interpolation schemes on the simulation of two-phase flows. In order to stabilize a numerical solution and assure a high numerical accuracy, the second-order upwind scheme is implemented into the CUPID code in the present paper. Some numerical tests have been performed with the implemented scheme and the comparison results between the second-order and first-order upwind schemes are introduced in the present paper. The comparison results among the two interpolation schemes and either the exact solutions or the mesh convergence studies showed the reduced numerical diffusion with the second order scheme.

  • PDF

Preparation and Characterization of Nylon 6,5 Copolymers from ε-Caprolactam and 2-Piperidone (ε-Caprolactam과 2-Piperidone으로부터 나일론 6,5 공중합체 제조 및 특성평가)

  • Kim, Hye Young;Goh, Jin-Seon;Ryu, Mi Hee;Kim, Dae Su;Song, Bong-Keun;Lee, Seung Hwan;Park, Si-Jae;Jegal, Jonggeon
    • Polymer(Korea)
    • /
    • v.38 no.1
    • /
    • pp.31-37
    • /
    • 2014
  • To prepare biomass based nylon 6,5 copolymers, ${\varepsilon}$-caprolactam and 2-piperidone, the monomers of nylon 6,5 copolymers, were synthesized respectively from lysine and 5-aminovaleric acid which were produced from glucose by the fermentation process. The copolymers were then polymerized by the anionic ring opening polymerization of them at $40^{\circ}C$, using potassium tert-butoxide as a catalyst and acetyl-2-caprolactam and carbon dioxide as initiators. The prepared copolymers were characterized with various analytical methods: their viscosity molecular weight ($M_{\eta}$) was as high as 30000 g/mol and polymerization yield was over 50%, and it was found that they were semi-crystalline polymers having melting point at $165^{\circ}C$ which was much lower than its thermal degradation point, $250^{\circ}C$. These polymers were expected to have good thermal processability and application fields.

STRONG CONVERGENCE OF COMPOSITE IMPLICIT ITERATIVE PROCESS FOR A FINITE FAMILY OF NONEXPANSIVE MAPPINGS

  • Gu, Feng
    • East Asian mathematical journal
    • /
    • v.24 no.1
    • /
    • pp.35-43
    • /
    • 2008
  • Let E be a uniformly convex Banach space and K be a nonempty closed convex subset of E. Let ${\{T_i\}}^N_{i=1}$ be N nonexpansive self-mappings of K with $F\;=\;{\cap}^N_{i=1}F(T_i)\;{\neq}\;{\theta}$ (here $F(T_i)$ denotes the set of fixed points of $T_i$). Suppose that one of the mappings in ${\{T_i\}}^N_{i=1}$ is semi-compact. Let $\{{\alpha}_n\}\;{\subset}\;[{\delta},\;1-{\delta}]$ for some ${\delta}\;{\in}\;(0,\;1)$ and $\{{\beta}_n\}\;{\subset}\;[\tau,\;1]$ for some ${\tau}\;{\in}\;(0,\;1]$. For arbitrary $x_0\;{\in}\;K$, let the sequence {$x_n$} be defined iteratively by $\{{x_n\;=\;{\alpha}_nx_{n-1}\;+\;(1-{\alpha}_n)T_ny_n,\;\;\;\;\;\;\;\;\; \atop {y_n\;=\;{\beta}nx_{n-1}\;+\;(1-{\beta}_n)T_nx_n},\;{\forall}_n{\geq}1,}$, where $T_n\;=\;T_{n(modN)}$. Then {$x_n$} convergence strongly to a common fixed point of the mappings family ${\{T_i\}}^N_{i=1}$. The result presented in this paper generalized and improve the corresponding results of Chidume and Shahzad [C. E. Chidume, N. Shahzad, Strong convergence of an implicit iteration process for a finite family of nonexpansive mappings, Nonlinear Anal. 62(2005), 1149-1156] even in the case of ${\beta}_n\;{\equiv}\;1$ or N=1 are also new.

  • PDF

A Piezo-Driven Miniaturized XY Stage with Two Prismatic-Prismatic Joints Type Parallel Compliant Mechanism (2 개의 병진-병진 관절형 병렬 탄성 메커니즘을 갖는 압전구동 소형 XY 스테이지)

  • Choi, Kee-Bong;Lee, Jae Jong;Kim, Gee Hong;Lim, Hyung Jun
    • Journal of the Korean Society for Precision Engineering
    • /
    • v.30 no.12
    • /
    • pp.1281-1286
    • /
    • 2013
  • In this paper, a miniaturized stage with two prismatic-prismatic joints (2-PP) type parallel compliant mechanism driven by piezo actuators is proposed. This stage consists of two layers which are a motion guide layer and an actuation layer. The motion guide layer has 2-PP type parallel compliant mechanism to guide two translational motions, whereas the actuation layer has two leverage type amplification mechanisms and two piezo actuators to generate forces. Since the volume of the stage is too small to mount displacement sensors, the piezo actuators embedding strain gauge sensors are chosen. With the strain gauge-embedded piezo actuators, a semi-control is implemented, which results in hysteresis compensation of the stage. As the results, the operating range of $30{\mu}m$, the resolution of 20 nm, and the bandwidth of 400 Hz in each axis were obtained in the experiments.

The convergence study on the Sexual life Experience of Rectal Cancer Men (직장암 남성의 성생활 경험에 관한 융합적 연구)

  • Woo, Sang-Jun;Lee, Eun-Sook
    • Journal of the Korea Convergence Society
    • /
    • v.8 no.5
    • /
    • pp.233-241
    • /
    • 2017
  • The purpose of this study is to understand the sex life of men with rectal cancer. Participants were 8 males who were treated with rectal cancer and analyzed semi-structured questionnaire by Downe-Wamboldt's content analysis method(Step 8). As a result of the analysis, three categories and nine subcategories were derived: Physical difficulty, Psychological difficulty, Mutual role difficulty. Thus, men treated with rectal cancer experience physical, psychological, and interrelated difficulties related to sex. This study was meaningful in that it revealed the experience related to sexual experience of men with rectal cancer and could be used as useful data for the development of sex - related programs.

SWIR-LWIR Photoluminescence from Sb-based Epilayers Grown on GaAs Substrates by using MBE

  • Hussain, Laiq;Pettersson, Hakan;Wang, Qin;Karim, Amir;Anderson, Jan;Jafari, Mehrdad;Song, Jindong;Choi, Won Jun;Han, Il Ki;Lim, Ju Young
    • Journal of the Korean Physical Society
    • /
    • v.73 no.11
    • /
    • pp.1604-1611
    • /
    • 2018
  • Utilizing Sb-based bulk epilayers on large-scale low-cost substrates such as GaAs for fabricating infrared (IR) photodetectors is presently attracting significant attention worldwide. For this study, three sample series of $GaAs_xSb_{1-x}$, $In_{1-x}Ga_xSb$, and $InAs_xSb_{1-x}$ with different compositions were grown on semi-insulating GaAs substrates by using molecular beam epitaxy (MBE) and appropriate InAs quantum dots (QDs) as a defect-reduction buffer layer. Photoluminescence (PL) signals from these samples were observed over a wide IR wavelength range from $2{\mu}m$ to $12{\mu}m$ in agreement with the expected bandgap, including bowing effects. In particular, interband PL signals from $InAs_xSb_{1-x}$ and $In_{1-x}Ga_xSb$ samples even at room temperature show promising potential for IR photodetector applications.