• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2001년도 가을 학술발표회 논문집
• /
• pp.201-210
• /
• 2001

In this study, large-scale distributed design approach for a life cycle cost (LCC) optimization of steel box girder bridges was implemented. A collaborative optimization approach is one of the multidisciplinary design optimization approaches and it has been proven to be best suited for distributed design environment. The problem of optimum LCC design of steel box girder bridges is formulated as that of minimization of the expected total LCC that consists of initial cost maintenance cost expected retrofit costs for strength, deflection and crack. To discuss the possibility of the application for the collaborative optimization of steel box girder bridges, the results of this algorithm are compared with those of single level algorithm. From the numerical investigations, the collaborative optimization approach proposed in this study may be expected to be new concepts and design methodologies associated with the LCC approach.

• 한국CDE학회논문집
• /
• 제5권3호
• /
• pp.263-275
• /
• 2000

Multidisciplinary design optimization(MDO) can yield optimal design considering all the disciplinary requirements concurrently. A method to implement the collaborative optimization(CO) approach, one of the MDO methodologies, is developed using a pre-compiler “EzpreCompiler”, a design optimization library “EzOptimizer”, and a common object request broker architecture(CORBA) in distributed computing environment. The CO approach is applied to a mathematical example to show its applicability and equivalence to standard optimization(SO) formulation. In a realistic engineering problem such as optimal design of a two-member hub frame, optimal design of a speed reducer and initial design of a bulk carrier, the CO yields better results than the SO. Furthermore, the CO allows the distributed processing using the CORBA, which leads to reduction of overall computation time.

• Journal of Mechanical Science and Technology
• /
• 제20권1호
• /
• pp.133-146
• /
• 2006

Collaborative optimization (CO) is a multi-level decomposed methodology for a large-scale multidisciplinary design optimization (MDO). CO is known to have computational and organizational advantages. Its decomposed architecture removes a necessity of direct communication among disciplines, guaranteeing their autonomy. However, CO has several problems at convergence characteristics and computation time. In this study, such features are discussed and some suggestions are made to improve the performance of CO. Only for the system level optimization, genetic algorithm is used and gradient-based method is used for subspace optimizers. Moreover, response surface models are replaced as analyses in subspaces. In this manner, CO is applied to aero-structural design problems of the aircraft wing and its results are compared with the multidisciplinary feasible (MDF) method and the original CO. Through these results, it is verified that the suggested approach improves convergence characteristics and offers a proper solution.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권7호
• /
• pp.1749-1773
• /
• 2024

Recent advances in deep neural networks (DNNs) have greatly improved the accuracy and universality of various intelligent applications, at the expense of increasing model size and computational demand. Since the resources of end devices are often too limited to deploy a complete DNN model, offloading DNN inference tasks to cloud servers is a common approach to meet this gap. However, due to the limited bandwidth of WAN and the long distance between end devices and cloud servers, this approach may lead to significant data transmission latency. Therefore, device-edge collaborative inference has emerged as a promising paradigm to accelerate the execution of DNN inference tasks where DNN models are partitioned to be sequentially executed in both end devices and edge servers. Nevertheless, collaborative inference in heterogeneous edge environments with multiple edge servers, end devices and DNN tasks has been overlooked in previous research. To fill this gap, we investigate the optimization problem of collaborative inference in a heterogeneous system and propose a scheme CIS, i.e., collaborative inference scheme, which jointly combines DNN partition, task offloading and scheduling to reduce the average weighted inference latency. CIS decomposes the problem into three parts to achieve the optimal average weighted inference latency. In addition, we build a prototype that implements CIS and conducts extensive experiments to demonstrate the scheme's effectiveness and efficiency. Experiments show that CIS reduces 29% to 71% on the average weighted inference latency compared to the other four existing schemes.

• 정보기술응용연구
• /
• 제1권
• /
• pp.173-211
• /
• 1999

The task of classification permeates all walks of life, from business and economics to science and public policy. In this context, nonlinear techniques from artificial intelligence have often proven to be more effective than the methods of classical statistics. The objective of knowledge discovery and data mining is to support decision making through the effective use of information. The automated approach to knowledge discovery is especially useful when dealing with large data sets or complex relationships. For many applications, automated software may find subtle patterns which escape the notice of manual analysis, or whose complexity exceeds the cognitive capabilities of humans. This paper explores the utility of a collaborative learning approach involving integrated models in the preprocessing and postprocessing stages. For instance, a genetic algorithm effects feature-weight optimization in a preprocessing module. Moreover, an inductive tree, artificial neural network (ANN), and k-nearest neighbor (kNN) techniques serve as postprocessing modules. More specifically, the postprocessors act as second0order classifiers which determine the best first-order classifier on a case-by-case basis. In addition to the second-order models, a voting scheme is investigated as a simple, but efficient, postprocessing model. The first-order models consist of statistical and machine learning models such as logistic regression (logit), multivariate discriminant analysis (MDA), ANN, and kNN. The genetic algorithm, inductive decision tree, and voting scheme act as kernel modules for collaborative learning. These ideas are explored against the background of a practical application relating to financial fraud management which exemplifies a binary classification problem.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4453-4468
• /
• 2015

This paper investigates optimization-based base station (BS) placement. An optimization model is defined and the BS placement problem is transformed to a lexicographical stratified programming (LSP) model for a given trajectory, according to different accuracy requirements. The feasible region for BS deployment is obtained from the positioning system requirement, which is also solved with signal coverage problem in BS placement. The LSP mathematical model is formulated with the average geometric dilution of precision (GDOP) as the criterion. To achieve an optimization solution, a tolerant factor based complete stratified series approach and grid searching method are utilized to obtain the possible optimal BS placement. Because of the LSP model utilization, the proposed algorithm has wider application scenarios with different accuracy requirements over different trajectory segments. Simulation results demonstrate that the proposed algorithm has better BS placement result than existing approaches for a given trajectory.

• 한국정밀공학회지
• /
• 제20권9호
• /
• pp.142-150
• /
• 2003

MDO is one of the efficient methods for huge and multi -functional system design. This paper describes a design collaboration framework with MDO in networked design environment. A prototype of web -based integrated design system was implemented to show sharing and exchange of models and analysis information between MDO modules and collaborative design stations. Server System consists of MDO modules for optimization and modeling module for 3D modeling operation. Client system provide user with graphic interface for shape modeling and system operation. We believe that the proposed approach can be extended to solve real complex multidisciplinary design problems.

• Journal of Hospice and Palliative Care
• /
• 제27권2호
• /
• pp.77-81
• /
• 2024

This case report explores the challenges and complexities associated with opioid management of cancer pain, emphasizing the importance of early involvement of a hospice consultation team and the adoption of a multidisciplinary approach to care. A 56-year-old man with advanced pancreatic cancer experienced escalating pain and inappropriate opioid prescriptions, highlighting the shortcomings of traditional pain management approaches. Despite procedural intervention by the attending physician and increased opioid dosages, the patient's condition deteriorated. Subsequently, the involvement of a hospice consultation team, in conjunction with collaborative psychiatric care, led to an overall improvement. The case underscores the necessity of early hospice engagement, psychosocial assessments, and collaborative approaches in the optimization of patient-centered palliative care.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권5호
• /
• pp.2381-2399
• /
• 2019

Traditional recommendation algorithms on Collaborative Filtering (CF) mainly focus on the rating prediction with explicit ratings, and cannot be applied to the top-N recommendation with implicit feedbacks. To tackle this problem, we propose a new collaborative filtering approach namely Maximize MAP with Matrix Factorization (MFMAP). In addition, in order to solve the problem of non-smoothing loss function in learning to rank (LTR) algorithm based on pairwise, we also propose a smooth MAP measure which can be easily implemented by standard optimization approaches. We perform experiments on three different datasets, and the experimental results show that the performance of MFMAP is significantly better than other recommendation approaches.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 추계학술발표대회
• /
• pp.442-445
• /
• 2023

최근 딥러닝 및 로봇기술의 발전으로 인해 대량의 데이터를 빠르게 수집하고 처리하는 연구 분야들로 확대되었다. 이와 관련된 한 가지 분야로써 다중 로봇을 이용한 분산학습 연구가 있으며, 이는 단일 에이전트를 이용할 때보다 대량의 데이터를 빠르게 수집 및 처리하는데 용이하다. 본 연구에서는 기존 Distributed Neural Network Optimization (DiNNO) 알고리즘에서 제안한 정적 분산 학습방법과 달리 단계적 분산학습 방법을 새롭게 제안하였으며, 모델 성능을 향상시키기 위해 원시 변수를 근사하는 단계수를 상수로 고정하는 기존의 방식에서 통신회차가 늘어남에 따라 점진적으로 근사 횟수를 높이는 방법을 고안하여 새로운 알고리즘을 제안하였다. 기존 알고리즘과 제안된 알고리즘의 정성 및 정량적 성능 평가를 수행하기 MNIST 분류와 2 차원 평면도 지도화 실험을 수행하였으며, 그 결과 제안된 알고리즘이 기존 DiNNO 알고리즘보다 동일한 통신회차에서 높은 정확도를 보임과 함께 전역 최적점으로 빠르게 수렴하는 것을 입증하였다.