强化学习
计算机科学
深度学习
数学优化
最优控制
整数规划
线性规划
模型预测控制
整数(计算机科学)
人工神经网络
代表(政治)
分段线性函数
构造(python库)
人工智能
控制(管理)
数学
算法
政治
政治学
程序设计语言
法学
几何学
作者
Vinicius Lima,Dzung T. Phan,Lam M. Nguyen,Jayant Kalagnanam
标识
DOI:10.23919/acc55779.2023.10155810
摘要
Deep learning models are frequently used to capture relations between inputs and outputs and to predict operation costs in dynamical systems. Computing optimal control policies based on the resulting regression models, however, is a challenging task because of the nonlinearity and nonconvexity of deep learning architectures. To address this issue, we propose in this paper a linearizable approach to design optimal control policies based on deep learning models for handling both continuous and discrete action spaces. When using piecewise linear activation functions, one can construct an equivalent representation of recurrent neural networks in terms of a set of mixed-integer linear constraints. That in turn means that the optimal control problem reduces to a mixed-integer linear program (MILP), which can then be solved using offthe-shelf MILP optimization solvers. Numerical experiments on standard reinforcement learning benchmarks attest to the good performance of the proposed approach.
科研通智能强力驱动
Strongly Powered by AbleSci AI