Smart Feasibility Pump: Reinforcement Learning for (Mixed) Integer Programming

no code implementations • 18 Feb 2021 • Meng Qi, Mengxin Wang, Zuo-Jun Shen

In addition to multi-layer perception (MLP), we propose a novel convolution neural network (CNN) structure for the policy network to capture the hidden information of the constraint matrix of the MIP problem.

reinforcement-learning Reinforcement Learning (RL)