Strong mixed-integer programming formulations for trained neural networks
نویسندگان
چکیده
منابع مشابه
Verifying Neural Networks with Mixed Integer Programming
Neural networks have demonstrated considerable success in a wide variety of real-world problems. However, the presence of adversarial examples slightly perturbed inputs that are misclassified with high confidence limits our ability to guarantee performance for these networks in safety-critical applications. We demonstrate that, for networks that are piecewise affine (for example, deep networks ...
متن کاملExtended Formulations in Mixed-Integer Convex Programming
We present a unifying framework for generating extended formulations for the polyhedral outer approximations used in algorithms for mixed-integer convex programming (MICP). Extended formulations lead to fewer iterations of outer approximation algorithms and generally faster solution times. First, we observe that all MICP instances from the MINLPLIB2 benchmark library are conic representable wit...
متن کاملStrong mixed-integer formulations for the floor layout problem
The floor layout problem (FLP) tasks a designer with positioning a collection of rectangular boxes on a fixed floor in such a way that minimizes total communication costs between the components. While several mixed integer programming (MIP) formulations for this problem have been developed, it remains extremely challenging from a computational perspective. This work takes a systematic approach ...
متن کاملMixed integer programming formulations for single machine scheduling problems
In this paper, the computational performance of four different mixed integer programming (MIP) formulations for various single machine scheduling problems is studied. Based on the computational results, we discuss which MIP formulation might work best for these problems. The results also reveal that for certain problems a less frequently used MIP formulation is computationally more efficient in...
متن کاملIncremental and encoding formulations for Mixed Integer Programming
The standard way to represent a choice between n alternatives in Mixed Integer Programming is through n binary variables that add up to one. Unfortunately, this approach commonly leads to unbalanced branch-and-bound trees and diminished solver performance. In this paper, we present an encoding formulation framework that encompasses and expands existing approaches to mitigate this behavior. Thro...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Mathematical Programming
سال: 2020
ISSN: 0025-5610,1436-4646
DOI: 10.1007/s10107-020-01474-5