On handling indicator constraints in mixed integer programming

113 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Predicting Tactical Solutions to Operational Planning Problems under Imperfect Information (2018, background): “(2017), Belotti et al. (2016), just to mention a few.”
Predictive Multiplicity in Classification (2019, background): “Althoughinteger programming is computationally intractable, modern solvers can solve empirical risk minimization for many real-world problems [see e.g., 6, 37, 47–49].”
Strong mixed-integer programming formulations for trained neural networks (2018, background): “ons presented in this paper also have connections with existing structures studied in the MIP and constraint programming communities, like indicator variables, on/oﬀ constraints, and convex envelopes [7,13,17,37,38,69]. Finally, this paper has connections with distributionally robust optimization, i”
Deep Neural Networks as 0-1 Mixed Integer Linear Programs: A Feasibility Study (2017, background): “It is known [1] that, for 0-1 MILP models like (4)–(7), the deﬁnition of tight upper bounds for the continuous variables appearing in the indicator constraints plays a crucial role for the practical resolution of the model itself.”
Abduction-Based Explanations for Machine Learning Models (2018, methodology): “To help with the encoding of ML models, we will exploit indicator constraints (e.g. (Belotti et al. 2016; Fischetti and Jo 2018)) of the form: where l i is some propositional literal.”