Neural Architecture Search for Joint Optimization of Predictive Power and Biological Knowledge

TL;DR

BioNAS is a neural architecture search framework designed for biomedical research that optimizes predictive accuracy while integrating biological knowledge, revealing novel insights from genomic data.

Contribution

Introduction of knowledge dissimilarity functions in BioNAS to jointly optimize predictive power and biological knowledge in neural architecture search.

Findings

BioNAS uncovers novel biological knowledge in genomics data.

It effectively balances predictive accuracy with biological interpretability.

BioNAS is accessible for domain experts to incorporate prior knowledge into deep learning.

Abstract

We report a neural architecture search framework, BioNAS, that is tailored for biomedical researchers to easily build, evaluate, and uncover novel knowledge from interpretable deep learning models. The introduction of knowledge dissimilarity functions in BioNAS enables the joint optimization of predictive power and biological knowledge through searching architectures in a model space. By optimizing the consistency with existing knowledge, we demonstrate that BioNAS optimal models reveal novel knowledge in both simulated data and in real data of functional genomics. BioNAS provides a useful tool for domain experts to inject their prior belief into automated machine learning and therefore making deep learning easily accessible to practitioners. BioNAS is available at https://github.com/zj-zhang/BioNAS-pub.