A neural network-based distributional constraint learning methodology for mixed-integer stochastic optimization Articles
Overview
published in
- EXPERT SYSTEMS WITH APPLICATIONS Journal
publication date
- June 2023
volume
- 232
Digital Object Identifier (DOI)
full text
International Standard Serial Number (ISSN)
- 0957-4174
Electronic International Standard Serial Number (EISSN)
- 1873-6793
abstract
-
The use of machine-learning methods helps to improve decision-making in different fields. In particular, the
idea of bridging predictions (predictive models) and prescriptions (optimization problems) is gaining attention
within the scientific community. One of the main ideas to address this trade-off is the Constraint Learning (CL)
methodology, where the structure of the machine learning model can be treated as a set of constraints to be
embedded within the optimization problem, establishing the relationship between a direct decision variable
𝑥 and a response variable 𝑦. However, most CL approaches have focused on making point predictions, not
considering the statistical and external uncertainty faced in the modeling process. In this paper, we extend
the CL methodology to deal with uncertainty in the response variable 𝑦. The novel Distributional Constraint
Learning (DCL) methodology makes use of a piece-wise linearizable neural network-based model to estimate
the parameters of the conditional distribution of 𝑦 (dependent on decisions 𝑥 and contextual information),
which can be embedded within mixed-integer optimization problems. In particular, we formulate a stochastic
optimization problem by sampling random values from the estimated distribution by using a linear set of
constraints. In this sense, DCL combines both the predictive performance of the neural network method and
the possibility of generating scenarios to account for uncertainty within a tractable optimization model. The
behavior of the proposed methodology is tested in the context of electricity systems, where a Virtual Power
Plant seeks to optimize its operation, subject to different forms of uncertainty, and with price-responsive
consumers.
Classification
subjects
- Statistics
keywords
- stochastic optimization; constraint learning; distribution estimation; neural networks; mixed-integer optimization