연구성과
Using Privileged Information to Improve Prediction in Health Data: A Case Study
- 작성자관리자
- 작성일2023.09.07
- 조회수82
Jeongoh Jeong, Do Hyung Kwon, Min Joon So, Anita Raja
상세내용
BibTeX
- 저자 : Jeongoh Jeong, Do Hyung Kwon, Min Joon So, Anita Raja
- 논문명 : Using Privileged Information to Improve Prediction in Health Data: A Case Study
- 학회명 : NeurIPS 2019 Workshop on Information Theory and Machine Learning (ITML 2019)
- 발간년도 : 2019
- 발간년도 : 2019
- 발간월 : December
- 초록 : The Learning Using Privileged information (LUPI) paradigm introduces the con-cept of utilizing Privileged Information (PI) as additional information availableduring training but not testing. In this paper, we investigate the effect of the noveluse of future data as PI on the classification of health datasets. Specifically, weconduct an empirical study to determine the characteristics of the datasets wherePI can contribute to improvement in the performance and convergence speeds ofthe predictive models
첨부파일
- 405CCameraReadySubmission5C40.pdf (274.1 KB)