Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 43 - Statistical Genetics II – New Models for Complex Study Designs
Type: Contributed
Date/Time: Monday, August 3, 2020 : 10:00 AM to 2:00 PM
Sponsor: Section on Statistics in Genomics and Genetics
Abstract #313246
Title: Transfer Learning in Cross-Project Test
Author(s): Shan Zhang*
Companies: Michigan State University
Keywords: Transfer Learning; Deep Learning; Functional Data Analysis
Abstract:

There are numerous research projects in genomic and genetics. Some of them share the same gene or similar research objects. Due to the high cost of research experiments, we tried to apply transfer learning to utilize the knowledge of similar research works to help one's own research project in this paper.

Transfer learning is a widely used method in computer science, which transfer the knowledge in one mature studies to a new one. Based on deep learning method, we can find complex features by training on related data, and help our topic with the well-tuned features.

We tested the effect of transferring knowledge of white people to black people at first. Then, we would investigate whether transfer learning is helpful for mice studies with the experimental data of rats.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program