[PDF][PDF] Learning with and for discrete optimization
MB Paulus - 2023 - research-collection.ethz.ch
Machine learning and discrete optimization are pillars of computer science and both are
widely used tools for analysis, prediction and decision-making across business, science and …
widely used tools for analysis, prediction and decision-making across business, science and …
Learning to Branch with Offline Reinforcement Learning
S Feng, Y Yang - 2023 - openreview.net
Mixed Integer Linear Program (MILP) solvers are mostly built upon a branch-and-bound
(B\&B) algorithm, where the efficiency of traditional solvers heavily depends on hand-craft …
(B\&B) algorithm, where the efficiency of traditional solvers heavily depends on hand-craft …
[PDF][PDF] Instance-specific algorithm configuration via unsupervised deep graph clustering.(2023)
ABSTRACT Instance-specific Algorithm Configuration (AC) methods are effective in
automatically generating high-quality algorithm parameters for heterogeneous NP-hard …
automatically generating high-quality algorithm parameters for heterogeneous NP-hard …
[PDF][PDF] IPGPT: Solving Integer Programming Problems with Sequence to Contrastive Multi-Label Learning
S Kong, C Liu, C Gomes - 2023 - cs.cornell.edu
Integer Programming (IP) is an essential class of combinatorial optimization problems
(COPs). Its inherent NP-hardness has fostered considerable efforts towards the …
(COPs). Its inherent NP-hardness has fostered considerable efforts towards the …
ILP-FORMER: Solving Integer Linear Programming with Sequence to Multi-Label Learning
S Kong, C Liu, CP Gomes - The 40th Conference on Uncertainty in … - openreview.net
Integer Linear Programming (ILP) is an essential class of combinatorial optimization
problems (COPs). Its inherent NP-hardness has fostered considerable efforts towards the …
problems (COPs). Its inherent NP-hardness has fostered considerable efforts towards the …
Empowering distributed constraint optimization with deep learning
Y Deng - 2023 - dr.ntu.edu.sg
Distributed Constraint Optimization Problems (DCOPs) are a fundamental formalism for multi-
agent coordination, in which a set of autonomous agents cooperatively find assignments to …
agent coordination, in which a set of autonomous agents cooperatively find assignments to …
A Symmetry-Aware Learning Approach for Solving Mixed-Integer Linear Programs
Q Chen, T Zhang, L Yang, Q Han, A Wang, R Sun… - openreview.net
Recently, machine learning techniques have been widely utilized for solving mixed-integer
linear programs (MILPs). Notably, learning-based approaches that encode MILPs as …
linear programs (MILPs). Notably, learning-based approaches that encode MILPs as …
BTBS-LNS: A Binarized-Tightening, Branch and Search Approach of Learning Large Neighborhood Search Policies for MIP
Learning to solve large-scale Mixed Integer Program (MIP) problems is an emerging
research topic, and policy learning-based Large Neighborhood Search (LNS) has recently …
research topic, and policy learning-based Large Neighborhood Search (LNS) has recently …