• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.234-237
• /
• 2005

강화학습을 위한 많은 방법 중 정책 반복을 이용한 actor-critic 학습 방법이 많은 적용 사례를 통해서 그 가능성을 인정받고 있다. Actor-critic 학습 방법은 제어입력 선택 전략을 위한 actor 학습과 가치 함수 근사를 위한 critic 학습이 필요하다. 본 논문은 critic의 학습을 위해 빠른 수렴성을 보장하는 RLS(recursive least square)를 사용하고, actor의 학습을 위해 정책의 기울기(policy gradient)를 이용하는 새로운 알고리즘을 제안하였다. 그리고 이를 실험적으로 확인하여 제안한 논문의 성능을 확인해 보았다.

• Journal of Korean Academy of Nursing
• /
• v.47 no.4
• /
• pp.467-475
• /
• 2017

Purpose: This study aimed to identify the actor and partner effects of self-efficacy, marital adjustment, and social support on the health promoting behavior of Korean pregnant couples. Methods: Participants were 132 couples who met the eligibility criteria. Data were collected from June to November, 2016 at a community health center. The Actor-Partner Interdependence Model was used for analyzing the actor and partner effects of self-efficacy, marital adjustment, and social support on health promoting behavior. Results: The fitness indices for the model were GFI=0.90, NFI=0.92, CFI=0.91, TLI=0.90, and RMSEA=0.04, which satisfied the criteria. Self-efficacy had actor and partner effect on health promoting behavior of wives, but had only actor effect of on health promoting behavior of husbands. Marital adjustment showed actor and partner effect on the health promoting behavior of pregnant couples. Social support only had an actor effect on the health promoting behavior of wives. And, marital adjustment and social support had a mutual effect. Conclusion: This study indicates that the partner involvement is needed to develop health promotion programs for pregnant couples.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.11 no.4
• /
• pp.267-274
• /
• 2011

Recently, actor-critic methods have drawn significant interests in the area of reinforcement learning, and several algorithms have been studied along the line of the actor-critic strategy. In this paper, we consider a new type of actor-critic algorithms employing the kernel methods, which have recently shown to be very effective tools in the various fields of machine learning, and have performed investigations on combining the actor-critic strategy together with kernel methods. More specifically, this paper studies actor-critic algorithms utilizing the kernel-based least-squares estimation and policy gradient, and in its critic's part, the study uses a sliding-window-based kernel least-squares method, which leads to a fast and efficient value-function-estimation in a nonparametric setting. The applicability of the considered algorithms is illustrated via a robot locomotion problem and a tunnel ventilation control problem.

• Journal of Korean Academy of Nursing
• /
• v.48 no.2
• /
• pp.154-166
• /
• 2018

Purpose: This study aimed to identify the actor and partner effects of health status, marital satisfaction, and self efficacy on retirement preparation in middle aged couples using the Actor-Partner Interdependence Model. Methods: Participants were 121 middle aged couples living in Seoul, Gyeonggi-do and Daegu City, Korea. All measures were self-administered. The collected data were analyzed using SPSS 20.0 and AMOS 20.0. Results: 1) Wife's self efficacy and marital satisfaction showed direct actor and partner effects on retirement preparation in middle aged couples. Wife's health status showed an indirect actor effect through wife's self efficacy and marital satisfaction on wife's retirement preparation, and showed an indirect partner effect through husband's self efficacy on husband's retirement preparation. 2) Husband's self efficacy and marital satisfaction showed only direct actor effect on retirement preparation. Health status among the husbands showed an indirect actor effect on husband's retirement preparation through their self efficacy and marital satisfaction, and an indirect partner effect through wife's marital satisfaction on wife's retirement preparation. 3) The actor effect size of health status on self efficacy and marital satisfaction was larger among husband's than wife's. Additionally, the partner effect size of health status on self efficacy and marital satisfaction was different between the two groups; however, the actor and partner effect size of self efficacy and marital satisfaction on retirement preparation was the same for husband and wife. Conclusion: The results of this study indicate that nursing intervention strategies for improving self efficacy and marital satisfaction are strongly needed to enhance retirement preparation of middle aged couples in Korean community.

• The Korean Journal of Community Living Science
• /
• v.27 no.spc
• /
• pp.557-571
• /
• 2016

This study examined the effects of the actor and partner according to the activity immersion conflict perceived by rural multicultural couples in terms of the self-esteem and happiness using Actor-Partner Interdependence Model (APIM). For this purpose, the data of 482 persons from 241 husband-immigrant women dyads were analyzed. As results of the analysis, only the effect of the actor among the actor and partner effect according to the activity immersion conflict affected the self-esteem significantly in both husband and immigrant women. Conflict, namely the actor and partner effect, affected the happiness in both the husband and immigrant women significantly. Second, the difference in the actor conflict level on activity immersion perceived by the couple according to the women's native country was significant in both husband and immigrant women. The level of actor conflict was highest in Vietnamese husbands and the Philippine women. On the other hand, the difference in the partner conflict level on the activity immersion according to the women's native country was significant only in the women, and the level of Vietnamese women was highest. Therefore, the characteristics in the level of actor and partner conflict on activity immersion was caused by the difference in the number of activity organizations participating between the husband and women. Based on the results, this study suggests some measures to reduce the conflict of couples through activity. In addition, some plans for a follow-up study using APIM are suggested.

• Korean Journal of Child Studies
• /
• v.38 no.2
• /
• pp.37-50
• /
• 2017

Objective: The purpose of this study was to assess actor and partner effects of mothers' and fathers' parenting stress and family cohesion and flexibility on warm parenting behavior. Methods: The sample consisted of 1,471 couples from the Panel Study of Korean Children. The sample was drawn from the sixth wave of data collection, including mothers and fathers of children aged 60-66 months. Data were analyzed via actor-partner interdependence models. Results: The findings illustrated that the actor effect of parenting stress on warm parenting behavior was significant in mothers and fathers, but there was no partner effect. They also revealed that actor and partner effects were significant in relation to mothers' and father's parenting stress and family cohesion and flexibility. Moreover, there was an actor effect of family cohesion on warm parenting behavior for mothers and fathers, but its partner effect was significant only in mothers. Finally, the mediating effects of family cohesion and flexibility between parenting stress and warm parenting behavior was verified. Conclusion: The results highlight not only the need for dyadic data analysis for mothers and fathers but also the need to consider in depth the fathers' role in parenting. Additional implications are discussed.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.16 no.4
• /
• pp.270-275
• /
• 2016

We present an approach for acceleration actor-critic algorithm for reinforcement learning with continuous action space. Actor-critic algorithm has already proved its robustness to the infinitely large action spaces in various high dimensional environments. Despite that success, the main problem of the actor-critic algorithm remains the same-speed of convergence to the optimal policy. In high dimensional state and action space, a searching for the correct action in each state takes enormously long time. Therefore, in this paper we suggest a search accelerating function that allows to leverage speed of algorithm convergence and reach optimal policy faster. In our method, we assume that actions may have their own distribution of preference, that independent on the state. Since in the beginning of learning agent act randomly in the environment, it would be more efficient if actions were taken according to the some heuristic function. We demonstrate that heuristically-accelerated actor-critic algorithm learns optimal policy faster, using Educational Process Mining dataset with records of students' course learning process and their grades.

• Journal of IKEEE
• /
• v.25 no.4
• /
• pp.644-649
• /
• 2021

Self-Imitation Learning is a simple off-policy actor-critic algorithm that makes an agent find an optimal policy by using past good experiences. In case that Self-Imitation Learning is combined with reinforcement learning algorithms that have actor-critic architecture, it shows performance improvement in various game environments. However, its applications are limited to reinforcement learning algorithms that have actor-critic architecture. In this paper, we propose a method of applying Self-Imitation Learning to Deep Q-Network which is a value-based deep reinforcement learning algorithm and train it in various game environments. We also show that Self-Imitation Learning can be applied to Deep Q-Network to improve the performance of Deep Q-Network by comparing the proposed algorithm and ordinary Deep Q-Network training results.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.893-898
• /
• 2005

Due to the merits that only a small amount of computation is needed for solutions and stochastic policies can be handled explicitly, the actor-critic algorithm, which is a class of reinforcement learning methods, has recently attracted a lot of interests in the area of artificial intelligence. The actor-critic network composes of tile actor network for selecting control inputs and the critic network for estimating value functions, and in its training stage, the actor and critic networks take the strategy, of changing their parameters adaptively in order to select excellent control inputs and yield accurate approximation for value functions as fast as possible. In this paper, we consider a new actor-critic algorithm employing an RLS(Recursive Least Square) method for critic learning, and policy gradients for actor learning. The applicability of the considered algorithm is illustrated with experiments on the two linked robot arm.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.519-524
• /
• 2009

Recently, reinforcement learning methods have drawn much interests in the area of machine learning. Dominant approaches in researches for the reinforcement learning include the value-function approach, the policy search approach, and the actor-critic approach, among which pertinent to this paper are algorithms studied for problems with continuous states and continuous actions along the line of the actor-critic strategy. In particular, this paper focuses on presenting a method combining the so-called ACFRL(actor-critic fuzzy reinforcement learning), which is an actor-critic type reinforcement learning based on fuzzy theory, together with the RLS-NAC which is based on the RLS filters and natural actor-critic methods. The presented method is applied to a control problem for crawling robots, and some results are reported from comparison of learning performance.