Abstract:The multicut problem, also known as correlation clustering, is a classic combinatorial optimization problem that aims to optimize graph partitioning given only node (dis)similarities on edges. It serves as an elegant generalization for several graph partitioning problems and has found successful applications in various areas such as data mining and computer vision. However, the multicut problem with an exponentially large number of cycle constraints proves to be NP-hard, and existing solvers either suffer from exponential complexity or often give unsatisfactory solutions due to inflexible heuristics driven by hand-designed mechanisms. In this article, we propose a deep graph reinforcement learning method to solve the multicut problem within a combinatorial decision framework involving sequential edge contractions. The customized subgraph neural network adapts to the dynamically edge-contracted graph environment by extracting bilevel connected features from both contracted and original graphs. Our method can learn to infer feasible multicut solutions end-to-end toward optimization of the multicut objective in a data-driven manner. More specifically, by exploring the decision space adaptively, it implicitly gains heuristic knowledge from topological patterns of instances and thereby generates more targeted heuristics overcoming the short-sightedness inherent in the hand-designed ones. During testing, the learned heuristics iteratively contract graphs to construct high-quality solutions within polynomial time. Extensive experiments on synthetic and real-world multicut instances show the superiority of our method over existing combinatorial solvers, while also maintaining a certain level of out-of-distribution generalization ability.

Deep Reinforcement Learning of Graph Matching.

Learning Visually Aligned Semantic Graph for Cross-Modal Manifold Matching.

GSC: A Graph-Based Skill Composition Framework for Robot Learning

Revocable Deep Reinforcement Learning with Affinity Regularization for Outlier-Robust Graph Matching

Deep Latent Graph Matching.

Deep Probabilistic Graph Matching

Partial Gromov-Wasserstein Learning for Partial Graph Matching

Learnable Graph Matching: A Practical Paradigm for Data Association

Combinatorial Learning of Robust Deep Graph Matching: an Embedding Based Approach.

GLMNet: Graph Learning-Matching Networks for Feature Matching

Learning Combinatorial Embedding Networks for Deep Graph Matching

Deep Learning of Partial Graph Matching Via Differentiable Top-K

Joint Graph Learning and Matching for Semantic Feature Correspondence

Neural Graph Matching Network: Learning Lawler’s Quadratic Assignment Problem with Extension to Hypergraph and Multiple-graph Matching

Deep Graph Reinforcement Learning for Solving Multicut Problem

Learning for Graph Matching and Related Combinatorial Optimization Problems

Self-supervised Learning of Visual Graph Matching

Learnable Graph Matching: Incorporating Graph Partitioning with Deep Feature Learning for Multiple Object Tracking

A Deep Reinforcement Learning Agent for Geometry Online Tutoring

Multi-level Graph Matching Networks for Deep and Robust Graph Similarity Learning

Fast Detection of Maximum Common Subgraph Via Deep Q-Learning