Deep Reinforcement Learning for Stabilization of Large-Scale Probabilistic Boolean Networks

Abstract

The ability to direct a Probabilistic Boolean Network (PBN) to a desired state is important to applications such as targeted therapeutics in cancer biology. Reinforcement Learning (RL) has been proposed as a framework that solves a discrete-time optimal control problem cast as a Markov Decision Process. We focus on an integrative framework powered by a model-free deep RL method that can address different flavours of the control problem (e.g., with or without control inputs; attractor state or a subset of the state space as the target domain). The method is agnostic to the distribution of probabilities for the next state, hence it does not use the probability transition matrix. The time complexity is only linear on the time steps, or interactions between the agent (deep RL) and the environment (PBN), during training. Indeed, we explore the scalability of the deep RL approach to (set) stabilization of large-scale PBNs and demonstrate successful control on large networks, including a metastatic melanoma PBN with 200 nodes.

A survey on applications of the semi‐tensor product method in Boolean control networks with time delays

“…Moreover, from the prospective of practical application, the study of Boolean networks using big data and artificial intelligence has lower complexity and broader application prospects. The reinforcement learning, including deep reinforcement learning (DRL), double deep‐Q Learning‐based, has emerged to solve the controllability, 97 output tracking, 98 stabilization, 99 synchronization 100 of probabilistic Boolean networks (PBNs). Hence, the reinforcement learning method for studying TBCNs deserves further analysis. Some issues regarding TBCNs have not been fully resolved yet, for instance, disturbance decoupling, system decomposition, function perturbations, and so forth.…”