http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Actor-Critic Algorithm with Transition Cost Estimation
Denisov Sergey,Jee-Hyong Lee 한국지능시스템학회 2016 INTERNATIONAL JOURNAL of FUZZY LOGIC and INTELLIGE Vol.16 No.4
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.
Actor-Critic Algorithm with Transition Cost Estimation
Sergey, Denisov,Lee, Jee-Hyong Korean Institute of Intelligent Systems 2016 INTERNATIONAL JOURNAL of FUZZY LOGIC and INTELLIGE Vol.16 No.4
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.
Russia, China, and the concept of Indo-Pacific
Igor Denisov,Oleg Paramonov,Ekaterina Arapova,Ivan Safranchuk 한양대학교 아태지역연구센터 2021 Journal of Eurasian Studies Vol.12 No.1
The newly minted concept of the “Indo-Pacific Region” (IPR) is generally seen as a response by the United States and its allies to China’s growing influence in strategically important areas of the Pacific and Indian oceans. However, the view of IPR as a single (U.S.-led) anti-Beijing front is simplistic and misleading, obscuring a variety of approaches by the region’s states. New Delhi has a strong tradition of non-alignment, whereas Tokyo is more interested in rules that restrict unilateral actions not only by China but also by other regional players, including the United States. Australian business is very cautious about frictions in trade relations with China. Beijing views the growing military activity of the United States off its shores, including in the South China Sea, as a threat to regional stability. According to the authoritative Chinese sources, the Indo-Pacific strategy of Donald Trump is part of broader efforts to prevent China from becoming a dominant regional and global power. At the same time, the development of Association of Southeast Asian Nations’ (ASEAN) understanding of the Indo-Pacific region is less of a concern to Beijing, as the South-East Asian countries interested in balancing China and the United States are unlikely to fully join the fight against the “authoritarian threat.” As for Russia, it unequivocally rejects the military/power-based U.S. version of the IPR concept and is more amenable to flexible versions promoted by other players, such as Tokyo’s multilateral vision for the Indo-Pacific Region. In the end, the final response of Russia and China to IPR will thus be determined not only by U.S. actions but also by the behavior of other regional powers.