MEMOIR adds branch-local and global memory with a reflection step to tree search for LLM solver synthesis, reaching 96.7% solution validity and 7.3-point score gains over baselines on seven CO problems with lower run-to-run variance.
An efficient graph convolutional network technique for the travelling salesman problem
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
GOAL uses conditioned diffusion on relational graphs with typed edges to produce feasible multi-objective solutions for scheduling problems, reporting 100% feasibility and sub-0.2% MAPE on FSP, JSP, and FJSP up to 20 jobs.
A two-stage ML sparsifier for TSP candidate graphs combines alpha-Nearest and POPMUSIC for high recall then trains a model to cut density while preserving coverage across distance types and instance sizes up to 500.
Multipartite GNN learns MILP formulations of network interdiction to outperform baselines on bi-level combinatorial tasks.
Presents three new training procedures for regression trees that enforce convex output constraints at training time and validates them on synthetic and hierarchical time-series data.
HyperNS clusters TSP cities with a sparse heatmap, builds a hyper tour over supernodes, and restricts neighborhood search to hyper-tour-relevant edges to improve solution quality on large instances.
Supplementary results on 1-tree relaxation performance inside a GCN-augmented branch-and-bound solver for TSP.
citing papers explorer
-
Memory-Guided Tree Search with Cross-Branch Knowledge Transfer for LLM Solver Synthesis
MEMOIR adds branch-local and global memory with a reflection step to tree search for LLM solver synthesis, reaching 96.7% solution validity and 7.3-point score gains over baselines on seven CO problems with lower run-to-run variance.
-
GOAL: Graph-based Objective-Aligned Diffusion Solvers for Dynamic Multi-Objective Optimization
GOAL uses conditioned diffusion on relational graphs with typed edges to produce feasible multi-objective solutions for scheduling problems, reporting 100% feasibility and sub-0.2% MAPE on FSP, JSP, and FJSP up to 20 jobs.
-
Machine Learning for Two-Stage Graph Sparsification for the Travelling Salesman Problem
A two-stage ML sparsifier for TSP candidate graphs combines alpha-Nearest and POPMUSIC for high recall then trains a model to cut density while preserving coverage across distance types and instance sizes up to 500.
-
Network Interdiction Goes Neural
Multipartite GNN learns MILP formulations of network interdiction to outperform baselines on bi-level combinatorial tasks.
-
Output-Constrained Decision Trees
Presents three new training procedures for regression trees that enforce convex output constraints at training time and validates them on synthetic and hierarchical time-series data.
-
Empowering Targeted Neighborhood Search via Hyper Tour for Large-Scale TSP
HyperNS clusters TSP cities with a sparse heatmap, builds a hyper tour over supernodes, and restricts neighborhood search to hyper-tour-relevant edges to improve solution quality on large instances.
-
Supplementary Materials to Graph Convolutional Branch and Bound
Supplementary results on 1-tree relaxation performance inside a GCN-augmented branch-and-bound solver for TSP.
- Convex Compositional Reasoning Models