In Proximity of ReLU DNN, PWA Function, and Explicit MPC
Abstract
Rectifier (ReLU) deep neural networks (DNN) and their connection with piecewise affine (PWA) functions is analyzed. The paper is an effort to find and study the possibility of representing explicit state feedback policy of model predictive control (MPC) as a ReLU DNN, and vice versa. The complexity and architecture of DNN has been examined through some theorems and discussions. An approximate method has been developed for identification of input-space in ReLU net which results a PWA function over polyhedral regions. Also, inverse multiparametric linear or quadratic programs (mp-LP or mp-QP) has been studied which deals with reconstruction of constraints and cost function given a PWA function.
Keywords
Cite
@article{arxiv.2006.05001,
title = {In Proximity of ReLU DNN, PWA Function, and Explicit MPC},
author = {Saman Fahandezh-Saadi and Masayoshi Tomizuka},
journal= {arXiv preprint arXiv:2006.05001},
year = {2020}
}
Comments
Submitted to Conference on Decision and Control (CDC) 2020