A framework to shift basins of attraction of gene regulatory networks through batch reinforcement learning

作者：

Highlights：

• Considers the problem of shifting gene regulatory networks to desirable basins of attraction.

• Novel framework has been developed for intervening through batch reinforcement learning techniques.

• mSFQI uses information regarding previous observations to deal with partial observability.

• Results show that our framework decreases the number of interventions applied.

摘要

•Considers the problem of shifting gene regulatory networks to desirable basins of attraction.•Novel framework has been developed for intervening through batch reinforcement learning techniques.•mSFQI uses information regarding previous observations to deal with partial observability.•Results show that our framework decreases the number of interventions applied.

论文关键词：Reinforcement learning,Gene regulatory network,Basin of attraction

论文评审过程：Received 15 October 2019, Revised 23 March 2020, Accepted 31 March 2020, Available online 16 May 2020, Version of Record 6 June 2020.

论文官网地址：https://doi.org/10.1016/j.artmed.2020.101853