Learning Cancer Outcomes from Heterogeneous Genomic Data Sources: An Adversarial Multi-task Learning Approach

Abstract

Translating the high-dimensional data generated by genomic platforms into reliable predictions of clinical outcomes remains a critical challenge in realizing the promise of genomic medicine largely due to small number of independent samples. We show that neural networks can be trained to predict clinical outcomes using heterogeneous genomic data sources via multi-task learning and adversarial representation learning, allowing one to combine multiple cohorts and outcomes in training. Experiments demonstrate that the proposed method helps mitigate data scarcity and outcome censorship in cancer genomics learning problems.

Tasks

Reproductions