http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
중국의 생성형 인공지능 창작물에 관한 법적 지위와 시사점에 관한 연구
허성진 한국법학회 2023 법학연구 Vol.92 No.-
Generative Artificial Intelligence refers to artificial intelligence that generates results according to user demands. It possesses the ability to produce various creative works such as images, text, and code through learning from the original data. Generative AI can produce satisfactory outcomes with relatively simple instructions from the user, often given in the form of keywords or other data. As a result, numerous AI-generated creations have emerged on the internet, and in 2022, an AI-created work called "Space Opera Theater" won at the Colorado State Fair Art Competition. Regarding the products of generative AI, South Korea does not recognize copyright, limiting authorship to natural persons based on legal precedents. Similarly, China, through Article 2 of its copyright law, grants copyright only to works created by natural persons or legal entities. However, legal precedents in China acknowledge that if users contribute significantly to the creation of AI-generated works, these works may be eligible for copyright protection. The production of works through generative AI is continuously increasing, and discussions on the copyright issues of such creations persist. Nevertheless, South Korea currently does not recognize copyright for AI-generated works, necessitating legal discussions on this matter. Therefore, this paper analyzes Chinese copyright laws and precedents that, despite not universally recognizing copyright for AI-generated works, consider the level of human contribution in determining copyright eligibility. South Korea, although not currently recognizing AI-generated works for copyright protection, may have the potential to recognize copyright for derivative works when users modify and adapt the results. Following the Chinese example, it is suggested that recognition of copyright may extend or narrow based on the degree of human involvement in the creation process, aligning with the trends of the times.
허성진,김지용,Hur, S.J.,Kim, J.Y. 한국전자통신연구원 2021 전자통신동향분석 Vol.36 No.2
In this paper, we discussed the necessity and importance of introducing feature stores to establish a collaborative environment between data engineering work and data science work. We examined the technology trends of feature stores by analyzing the status of some major feature stores. Moreover, by introducing a feature store, we can reduce the cost of performing artificial intelligence (AI) projects and improve the performance and reliability of AI models and the convenience of model operation. The future task is to establish technical requirements for establishing a collaborative environment between data engineering work and data science work and develop a solution for providing a collaborative environment based on this.
류마티스관절염에서 tumor necrosis factor-α에 의한 윤활막 증식에 관여하는 유전자 탐색
허성진,윤지희 대한체질인류학회 2003 해부·생물인류학 (Anat Biol Anthropol) Vol.16 No.2
Screening of Genes Regulating TNF-α-mediated Synovial Hyperplasia in Rheumatoid Arthritis Sung-Jin Huh, Jeehee Youn
핸드오버시 재전송 부담 감소를 위한 데이터그램 전송 정책
허성진,김정삼,정재열,강상용,한기준,Heo, Seong-Jin,Kim, Jeong-Sam,Jeong, Jae-Yeol,Kang, Sang-Yong,Han, Ki-Jun 대한전자공학회 1999 電子工學會論文誌, S Vol.s36 No.1
본 논문에서는, 핸드오버시의 결로 재설정으로 이동 호스트에서 수신하는 패킷의 순서가 뒤바뀌고 이로 인하여 전송측에서 불필요한 재전송을 수행하게 되는 비효율성을 해결하기 위하여 홈 에이전트가 등록 요청 메시지를 수신하면 새로운 외부 에이전트로의 전송을 멈추고 버퍼링을 수행하다가 일정시간이 경과한 후 전송을 재개하는 방안을 제시하였다. 홈 에이전트에서의 데이터그램 전송 절차를 개선한 본 논문의 방안을 사용하여 불필요한 재전송으로 인한 종단간 성능 저하의 문제가 적은 비용으로 해결될 수 있음을 시뮬레이션을 통하여 확인하였다. In this paper, we propose a new datagram delivery strategy at the home agent to improve the end-to-end performance by reducing retransmission overheads during handover. Changed routing might be able to cause packet disordering and this in turn could cause unnecessary retransmission at the fixed host and finally results in performance degradation. In our proposal, the home agent begins to buffer received datagrams instead of transmitting them after receiving a registration request message from the foreign agent and then transmits them again after for a certain time. Simulation results show that our proposal may successfully solve this problem at little cost.