Structure-Based Inverse Reinforcement Learning for Quantification of Biological Knowledge
Structure-Based Inverse Reinforcement Learning for Quantification of Biological Knowledge
复制标题
DOI：
10.1109/cai54212.2023.00126
复制
发表时间：
2023-06
期刊：
影响因子：
--
通讯作者：
A. Ravari;Seyede Fatemeh Ghoreishi;Mahdi Imani
中科院分区：
文献类型：
--
作者：
A. Ravari;Seyede Fatemeh Ghoreishi;Mahdi Imani
基因调控网络（GRNs）在各种细胞过程中起着至关重要的作用，包括应激反应，DNA修复以及癌症等复杂疾病的机制。生物学家参与了大多数生物学分析。因此，量化反映在现有生物数据中的政策可以显着帮助我们更好地了解这些复杂的系统。阻止利用现有机器学习，特别是反向强化学习技术来量化生物学家知识的主要挑战是生物数据的局限性和巨大的不确定性。本文利用GRNs的网络结构来定义专家奖励函数，该函数包含的参数比常规奖励模型少。使用哺乳动物细胞周期和合成基因表达数据的数值实验表明，所提出的方法在量化生物学家的政策的上级性能。
Gene regulatory networks (GRNs) play crucial roles in various cellular processes, including stress response, DNA repair, and the mechanisms involved in complex diseases such as cancer. Biologists are involved in most biological analyses. Thus, quantifying their policies reflected in available biological data can significantly help us to better understand these complex systems. The primary challenges preventing the utilization of existing machine learning, particularly inverse reinforcement learning techniques, to quantify biologists’ knowledge are the limitations and huge amount of uncertainty in biological data. This paper leverages the network-like structure of GRNs to define expert reward functions that contain exponentially fewer parameters than regular reward models. Numerical experiments using mammalian cell cycle and synthetic gene-expression data demonstrate the superior performance of the proposed method in quantifying biologists’ policies.