• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2022년도 제34회 한글 및 한국어 정보처리 학술대회
• /
• pp.77-81
• /
• 2022

최근 사전학습 언어모델에 내재된 지식을 최대한으로 활용하고자 태스크에 대한 설명을 입력으로 주는 manual prompt tuning 방법과 자연어 대신 학습가능한 파라미터로 태스크에 대한 이해를 돕는 soft prompt tuning 방법론이 자연어처리 분야에서 활발히 연구가 진행되고 있다. 이에 본 연구에서는 페르소나 대화 생성 태스크에서 encoder-decoder 구조 기반의 사전학습 언어모델 BART를 활용하여 manual prompt tuning 및 soft prompt tuning 방법을 고안하고, 파인튜닝과의 성능을 비교한다. 전체 학습 데이터에 대한 실험 뿐 아니라, few-shot 세팅에서의 성능을 확인한다.

• 드라이브 ㆍ 컨트롤
• /
• 제13권3호
• /
• pp.15-23
• /
• 2016

The direct-drive servo valve(DDV) is a kind of one-stage valve because the main spool valve is directly driven by the dc motor. Since the DDV structure is simple, it is less expensive, more reliable, and offers a reduced internal leakage and a reduced sensitivity to fluid contamination. The control system of the DDV is highly nonlinear due to a current limiter, a voltage limiter, and the flow-force effect on the spool motion. The shape of the step response of the DDV-control system varies considerably according to the magnitudes of the step input and the load pressure. The system-design requirements mean that the overshoots should be less than 20%, and the errors at 0.02s should be less than 2%, regardless of the reference-step input sizes of 1V and 5V and the load-pressure magnitudes of 0MPa and 20.7MPa. To satisfy the system-design requirements, the PID-controller parameters of $K_c$, $T_i$ and $T_d$, and the derivative-feedback gain of $K_{der}$ are designed using the root locus and manual tuning.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 추계학술대회 논문집 학회본부A
• /
• pp.176-178
• /
• 1998

Recently, Control method of induction Motor is applied in full circuit model, as circular diagram. In Elevetor moderization problem, there is no circuit information. Nothing but, Motor terminal voltage and HP of motor. So, in this study, using KS induction Motor table, try to solve initial valve problem and make some implementations of manual tuning for use of automatic tuning of induction motor parameter.

• 자동차안전학회지
• /
• 제13권4호
• /
• pp.20-25
• /
• 2021

This paper is the result of a research on hybrid tuning technology developed to improve the actual fuel efficiency and reduce emissions of in-use light-duty diesel trucks. In this study, a hybrid powertrain was constructed by inserting an electric motor between the diesel engine and manual transmission of an internal combustion engine vehicle and installing a battery. To verify the safety, a test was conducted based on the Korean tuning regulations. In particular, since there has been no case of tuning an internal combustion engine vehicle into a hybrid vehicle in Korea, it was necessary to carry out all procedures to receive tuning approval. The approval process consists of a technical review, safety verification test, and application for tuning approval. As a result, the test vehicle was approved for tuning because both the technical review and vehicle test results were suitable. Therefore, this study confirmed the safety of diesel hybrid tuning technology, and laid the foundation for the research and development of technologies to tune into an eco-friendly vehicle as well as the activation of related industries.

• 제어로봇시스템학회논문지
• /
• 제6권10호
• /
• pp.888-893
• /
• 2000

This paper deals with the development of an automatic design software for DC servo motor control, which provides good performance with rapid response and velocity control accuracy. In the proposed method, the design is automatically executed using Matlab, and iterative learning control algorithms are used in the design process. We applied this method to 50W, 100W, 200W, 300W, 500W, 750W, 1.8kW and 4.5kW DC servo motors which are widely used in the industry. We compare the results of the manual tuning design method with that of the automatic design method presented in this paper. From the experimental results, we can find that the performance of the proposed method is better than that of the manual tuning design method.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2002년도 춘계학술대회 논문집
• /
• pp.621-624
• /
• 2002

The purpose of this paper is a study on getting proper gain set of PID controller which satisfies multi-performance specifications of the control system. The multi-objective optimization method is introduced to evaluate specifications, and the genetic algorithm is used as an optimal problem solver. To enhance the performance of genetic algorithm itself, adaptive technique is included. According to the proposed method in this paper, finding suitable gain set can be more easily accomplishable than manual gain seeking and tuning.

• 제어로봇시스템학회논문지
• /
• 제5권8호
• /
• pp.914-922
• /
• 1999

A study on the improvement of tracking performance of a 3 DOF planar parallel manipulator is performed. A class of adaptive tracking control sheme is designed using self tuning adaptive fuzzy logic control theory. This control sheme is composed of three classical PD controller and a multi learning type self tuning adaptive fuzzy logic controller set. PD controller is tuned roughly by manual setting a priori and fuzzy logic controller is tuned precisely by the gradient descent method for a global solution during run-time, so the proposed control scheme is tuned more rapidly and precisely than the single learning type self tuning adaptive fuzzy logic control sheme for a local solution. The control performance of the proposed algorithm is verified through experiments.

• International Journal of Control, Automation, and Systems
• /
• 제4권5호
• /
• pp.575-591
• /
• 2006

In a previous research, we presented the Grid Method and confirmed it as a systematic and efficient problem formulation method for the task-oriented design of robot kinematics. However, our previous research was limited in two ways. First, it gave only a local optimum due to its use of a local optimization technique. Second, it used constant weights for a cost function chosen by the manual weights tuning algorithm, thereby showing low efficiency in finding an optimal solution. To overcome these two limitations, therefore, this paper presents a global optimization technique and an adaptive weights tuning algorithm to solve a formulated problem using the Grid Method. The efficiencies of the proposed algorithms have been confirmed through the kinematic design examples of various robot manipulators.

• 한국컴퓨터정보학회논문지
• /
• 제24권1호
• /
• pp.137-143
• /
• 2019

As the performance of computers increases, the use of deep learning, which has faced technical limitations in the past, is becoming more diverse. In many fields, deep learning has contributed to the creation of added value and used on the bases of more data as the application become more divers. The process for obtaining a better performance model will require a longer time than before, and therefore it will be necessary to find an optimal model that shows the best performance more quickly. In the artificial neural network modeling a tuning process that changes various elements of the neural network model is used to improve the model performance. Except Gride Search and Manual Search, which are widely used as tuning methods, most methodologies have been developed focusing on heuristic algorithms. The heuristic algorithm can get the results in a short time, but the results are likely to be the local optimal solution. Obtaining a global optimal solution eliminates the possibility of a local optimal solution. Although the Brute Force Method is commonly used to find the global optimal solution, it is not applicable because of an infinite number of hyper parameter combinations. In this paper, we use a statistical technique to reduce the number of possible cases, so that we can find the global optimal solution.

• 한국HCI학회:학술대회논문집
• /
• 한국HCI학회 2009년도 학술대회
• /
• pp.443-448
• /
• 2009

본 논문에서는 필터링 기반으로 한 이미지 만화화를 위한 파라미터 조절을 위해 대화형 유전자 알고리즘을 이용한 시스템을 제안한다. 만화화의 스타일은 사람마다 주관적이고, 비전문가가 파라미터를 직접 조절하는 데에는 시그널 프로세싱에 대한 이해가 요구되므로 쉽지 않은 일이다. 우리는 이러한 문제점을 해결하기 위해 사용자에게 직접 평가함수를 받고 사용자가 원하는 방향으로 해를 찾아주는 대화형 유전자 알고리즘 기법을 이용하는 인터페이스 기술을 제안한다. 이 방법을 이용하면 비전문적인 사용자도 원하는 스타일의 만화화를 생성하는 파라미터를 비교적 빠른 시간 안에 설정해 줄 수 있었다.