首页|Reports Outline Machine Learning Findings from Agency for Science Technology and Research (A*STAR) (Reliability-improved Machine Learning Model Using Knowledge- embedded Learning Approach for Smart Manufacturing)
Reports Outline Machine Learning Findings from Agency for Science Technology and Research (A*STAR) (Reliability-improved Machine Learning Model Using Knowledge- embedded Learning Approach for Smart Manufacturing)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Investigators discuss new findings in Machine Learning. According to news originating from Singapore, Singapore, by Ne wsRx correspondents, research stated, "Machine learning models play a crucial ro le in smart manufacturing by revolutionizing industrial automation so as to boos t productivity and product quality. However, the reliability of these models oft en faces challenges from factors such as data drift, concept drift, adversarial attacks, and increasing model complexity." Our news journalists obtained a quote from the research from Agency for Science Technology and Research (A*STAR), "In addressing these challenges, this paper pr oposes a novel approach called Reliability Improved Machine Learning (RIML), whi ch leverages on prior knowledge by incorporating it into the machine learning pi peline through a secondary output that is easily verifiable and assessable withi n the application domain. Built upon the Knowledge-embedded Machine Learning (KM L) framework, RIML differs from conventional strategies by modifying the model's architecture. In its implementation, additional layers were introduced, specifi cally designed to identify and discard misclassified cases to improve the model' s reliability. RIML's efficacy was successfully demonstrated through a simulated dataset and three real use-case studies, namely, a general walk/run scenario, a n industry-related case using metro railway dataset, and a smart manufacturing a pplication on gas detection."
SingaporeSingaporeAsiaCyborgsEme rging TechnologiesMachine LearningAgency for Science Technology and Research (A*STAR)
NETL
NSTL
