• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1150-1156
• /
• 2019

This paper explores a model-free value-based approach for solving survival gridworld problem. Survival gridworld problem opens up a challenge involving taking risks to gain better rewards. Classic value-based approach in model-free reinforcement learning assumes minimal risk decisions. The proposed method involves a hybrid on-policy and off-policy updates to experience roll-outs using a modified Q-based update equation that introduces a parametric linear rectifier and motivational discount. The significance of this approach is it allows model-free training of agents that take into account risk factors and motivated exploration to gain better path decisions. Experimentations suggest that the proposed method achieved better exploration and path selection resulting to higher episode scores than classic off-policy and on-policy Q-based updates.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.13 no.5
• /
• pp.57-67
• /
• 2003

Security Policy Model is a structured representation using informal, semiformal or formal method of security policy to be enforced by TOE. It provides TOE to get an assurance to mitigate security flaws resulted from inconsistency between security functional requirements and functional specifications. Therefore, Security Policy Model has been required under an hish evaluation assurance level on an evaluation criteria such as ISO/IEC 15408(Common Criteria, CC). In this paper, we present an evaluation method for security policy model based on assurance requirements for security policy model in Common Criteria through an analysis of concepts, related researches and assurance requirements for security policy model.

• Journal of Korea Multimedia Society
• /
• v.23 no.7
• /
• pp.830-839
• /
• 2020

As a popular mathematical framework for modeling decision making, Markov decision process (MDP) has been widely used to solve problem in many engineering fields. MDP consists of a set of discrete states, a finite set of actions, and rewards received after reaching a new state by taking action from the previous state. The objective of MDP is to find an optimal policy, that is, to find the best action to be taken in each state to maximize the expected discounted reward of policy (EDR). In practice, MDP is typically unknown, so simulation-based policy improvement (SBPI), which improves a given base policy sequentially by selecting the best action in each state depending on rewards observed via simulation, can be a practical way to find the optimal policy. However, the efficiency of SBPI is still a concern since many simulation samples are required to precisely estimate EDR for each action in each state. In this paper, we propose a method to select the best action accurately in each state using a small number of simulation samples, thereby improving the efficiency of SBPI. The proposed method accumulates the simulation samples observed in the previous states, so it is possible to precisely estimate EDR even with a small number of samples in the current state. The results of comparative experiments on the existing method demonstrate that the proposed method can improve the efficiency of SBPI.

• Journal of the Korea Society of Computer and Information
• /
• v.8 no.4
• /
• pp.159-164
• /
• 2003

Location of stock in a warehouse directly affects the total materials handling expense of all goods moving through the warehouse. The purpose of this paper is to develop a storage policy in order picking warehouse, the block-based storage policy to minimize the total order picking time. In block-based storage policy, the rack is divided into blocks and items are assigned to each block based on the turn-over rate of each item and the average distance between the blocks and the dock. To demonstrate the performance of the proposed policy, we compare with the existing method called class-based storage policy under various matching methods.

• Journal of KIISE:Software and Applications
• /
• v.37 no.5
• /
• pp.363-369
• /
• 2010

Demonstration-based learning has the advantage that a user can easily teach his/her robot new task knowledge just by demonstrating directly how to perform the task. However, many previous demonstration-based learning techniques used a kind of attribute-value vector model to represent their state spaces and policies. Due to the limitation of this model, they suffered from both low efficiency of the learning process and low reusability of the learned policy. In this paper, we present a new demonstration-based learning method, in which the relational model is adopted in place of the attribute-value model. Applying the relational instance-based learning to the training examples extracted from the records of the user demonstrations, the method derives a relational instance-based policy which can be easily utilized for other similar tasks in the same domain. A relational policy maps a context, represented as a pair of (state, goal), to a corresponding action to be executed. In this paper, we give a detail explanation of our demonstration-based relational policy learning method, and then analyze the effectiveness of our learning method through some experiments using a robot simulator.

• The Journal of Fisheries Business Administration
• /
• v.49 no.4
• /
• pp.1-17
• /
• 2018

The Korean fisheries industry is a traditional business, the majority of which are small and medium-sized enterprises (SMEs). It has played an important role in the South Korean economies in the past several decades, but it currently faces the limitations of growth potential and profitability due to declining workforce, aging populations, deteriorating fishery environments, climate changes, and rapid changes in the global industrial ecosystem. Many studies have suggested solutions for the fisheries industry in macro perspective, but there are rarely any studies taking the strategic approaches for the problem. If it is possible for governments to support the companies that are likely to increase their value-added selectively, it will break through the current situation more effectively. This paper introduces a study on the selection method utilizing data envelopment analysis (DEA) to find SMEs with potentials to increase profits and growth. We suggest selecting SMEs with high management efficiency and ability to utilize intangible assets as the target companies. We also suggest policy objectives for SMEs in the domestic fisheries industry based on the results of DEA analysis and propose a data-based method for the policy decisions.

• Korean System Dynamics Review
• /
• v.15 no.2
• /
• pp.75-96
• /
• 2014

The purpose of this paper is to analyze the functional changes of local educational authority policy based on the systems thinking perspective using causal loop diagrams. In the past, the main function of the local educational authority was to manage and supervise schools. Through this policy, local educational authority would be transformed into a support agency. However, this policy did not achieve the goal, was to cause confusion and require improvement. This study shows structured causes of the problem based on systems thinking. These diagrams make it possible for educational policy makers to provide ideas, although they have some complicated environment. The findings indicate that based on systems thinking in this policy can help those who related to policy decision than existing diagnosis method.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.10
• /
• pp.73-81
• /
• 2017

In this paper, we propose an analysis framework to capture the trends of information security incidents and evaluate the security policy based on the incident analysis. We build a big data from news media collecting security incidents news and policy news, identify key trends in information security from this, and present an analytical method for evaluating policies from the point of view of incidents. In more specific, we propose a network-based analysis model that allows us to easily identify the trends of information security incidents and policy at a glance, and a cosine similarity measure to find important events from incidents and policy announcements.

• Journal of Information Technology Services
• /
• v.11 no.1
• /
• pp.83-92
• /
• 2012

The development of science and technology oriented knowledge society accelerates the convergence between scientific theory and industrial technology and increases the complexity problem of social and economic sectors. These cause the difficulty of securing the reliability and objectivity of science and technology policy. These also are barriers of balanced evaluation between rational science and technology policy making, management, and policy coordination. In this regard, Advanced countries in science and technology develops policy support system and promotes the program of evidence-based SciSIP(Science of Science and Innovation policy) together. This paper introduces a new approach developing science service of science and technology policy utilizing business intelligence technology in Korea. Also, it proposes the integration method of policy knowledge base and component-based service supporting S&T policy decision-making process and introduces services case studies.

• Health Policy and Management
• /
• v.10 no.2
• /
• pp.1-21
• /
• 2000

This study reviewde various parametic and nonparametic method for forexasting hospital closures in Korea. We compared multivariate discriminant analysis, multivartiate logistic regression, classfication and regression tree, and neural network method based on hit ratio of each model for forecasting hospital closure. Like other studies in the literture, neural metwork analysis showed highest average hit ratio. For policy and business purposes, we combined the four analytical method and constructed a foreasting model that can be easily used to predict the probabolity of hospital closure given financial information of a hospital.