Message passing neural networks (MPNNs) operate on graphs by exchanging information between neigbouring nodes. MPNNs have been successfully applied to various node-, edge-, and graph-level tasks in areas like molecular science, computer vision, natural language processing, and combinatorial optimization. However, most MPNNs require training on large amounts of labeled data, which can be costly and time-consuming. In this work, we explore the use of various untrained message passing layers in graph neural networks, i.e. variants of popular message passing architecture where we remove all trainable parameters that are used to transform node features in the message passing step. Focusing on link prediction, we find that untrained message passing layers can lead to competitive and even superior performance compared to fully trained MPNNs, especially in the presence of high-dimensional features. We provide a theoretical analysis of untrained message passing by relating the inner products of features implicitly produced by untrained message passing layers to path-based topological node similarity measures. As such, untrained message passing architectures can be viewed as a highly efficient and interpretable approach to link prediction.
Combinatorial Optimization is crucial to numerous real-world applications, yet still presents challenges due to its (NP-)hard nature. Amongst existing approaches, heuristics often offer the best trade-off between quality and scalability, making them suitable for industrial use. While Reinforcement Learning (RL) offers a flexible framework for designing heuristics, its adoption over handcrafted heuristics remains incomplete within industrial solvers. Existing learned methods still lack the ability to adapt to specific instances and fully leverage the available computational budget. The current best methods either rely on a collection of pre-trained policies, or on data-inefficient fine-tuning; hence failing to fully utilize newly available information within the constraints of the budget. In response, we present MEMENTO, an approach that leverages memory to improve the adaptation of neural solvers at inference time. MEMENTO enables updating the action distribution dynamically based on the outcome of previous decisions. We validate its effectiveness on benchmark problems, in particular Traveling Salesman and Capacitated Vehicle Routing, demonstrating its superiority over tree-search and policy-gradient fine-tuning; and showing it can be zero-shot combined with diversity-based solvers. We successfully train all RL auto-regressive solvers on large instances, and show that MEMENTO can scale and is data-efficient. Overall, MEMENTO enables to push the state-of-the-art on 11 out of 12 evaluated tasks.
We introduce iterative retrieval, a novel framework that empowers retrievers to make iterative decisions through policy optimization. Finding an optimal portfolio of retrieved items is a combinatorial optimization problem, generally considered NP-hard. This approach provides a learned approximation to such a solution, meeting specific task requirements under a given family of large language models (LLMs). We propose a training procedure based on reinforcement learning, incorporating feedback from LLMs. We instantiate an iterative retriever for composing in-context learning (ICL) exemplars and apply it to various semantic parsing tasks that demand synthesized programs as outputs. By adding only 4M additional parameters for state encoding, we convert an off-the-shelf dense retriever into a stateful iterative retriever, outperforming previous methods in selecting ICL exemplars on semantic parsing datasets such as CalFlow, TreeDST, and MTOP. Additionally, the trained iterative retriever generalizes across different inference LLMs beyond the one used during training.
Machine Learning-based heuristics have recently shown impressive performance in solving a variety of hard combinatorial optimization problems (COPs). However they generally rely on a separate neural model, specialized and trained for each single problem. Any variation of a problem requires adjustment of its model and re-training from scratch. In this paper, we propose GOAL (for Generalist combinatorial Optimization Agent Learning), a generalist model capable of efficiently solving multiple COPs and which can be fine-tuned to solve new COPs. GOAL consists of a single backbone plus light-weight problem-specific adapters for input and output processing. The backbone is based on a new form of mixed-attention blocks which allows to handle problems defined on graphs with arbitrary combinations of node, edge and instance-level features. Additionally, problems which involve heterogeneous types of nodes or edges are handled through a novel multi-type transformer architecture, where the attention blocks are duplicated to attend the meaningful combinations of types while relying on the same shared parameters. We train GOAL on a set of routing, scheduling and classic graph problems and show that it is only slightly inferior to the specialized baselines while being the first multi-task model that solves a wide range of COPs. Finally we showcase the strong transfer learning capacity of GOAL by fine-tuning it on several new problems. Our code is available at https://github.com/naver/goal-co/.
Recent years have witnessed a growing trend toward employing deep reinforcement learning (Deep-RL) to derive heuristics for combinatorial optimization (CO) problems on graphs. Maximum Coverage Problem (MCP) and its probabilistic variant on social networks, Influence Maximization (IM), have been particularly prominent in this line of research. In this paper, we present a comprehensive benchmark study that thoroughly investigates the effectiveness and efficiency of five recent Deep-RL methods for MCP and IM. These methods were published in top data science venues, namely S2V-DQN, Geometric-QN, GCOMB, RL4IM, and LeNSE. Our findings reveal that, across various scenarios, the Lazy Greedy algorithm consistently outperforms all Deep-RL methods for MCP. In the case of IM, theoretically sound algorithms like IMM and OPIM demonstrate superior performance compared to Deep-RL methods in most scenarios. Notably, we observe an abnormal phenomenon in IM problem where Deep-RL methods slightly outperform IMM and OPIM when the influence spread nearly does not increase as the budget increases. Furthermore, our experimental results highlight common issues when applying Deep-RL methods to MCP and IM in practical settings. Finally, we discuss potential avenues for improving Deep-RL methods. Our benchmark study sheds light on potential challenges in current deep reinforcement learning research for solving combinatorial optimization problems.
Active learning is increasingly adopted for expensive multi-objective combinatorial optimization problems, but it involves a challenging subset selection problem, optimizing the batch acquisition score that quantifies the goodness of a batch for evaluation. Due to the excessively large search space of the subset selection problem, prior methods optimize the batch acquisition on the latent space, which has discrepancies with the actual space, or optimize individual acquisition scores without considering the dependencies among candidates in a batch instead of directly optimizing the batch acquisition. To manage the vast search space, a simple and effective approach is the greedy method, which decomposes the problem into smaller subproblems, yet it has difficulty in parallelization since each subproblem depends on the outcome from the previous ones. To this end, we introduce a novel greedy-style subset selection algorithm that optimizes batch acquisition directly on the combinatorial space by sequential greedy sampling from the greedy policy, specifically trained to address all greedy subproblems concurrently. Notably, our experiments on the red fluorescent proteins design task show that our proposed method achieves the baseline performance in 1.69x fewer queries, demonstrating its efficiency.
We introduce a new method to jointly reduce the dimension of the input and output space of a function between high-dimensional spaces. Choosing a reduced input subspace influences which output subspace is relevant and vice versa. Conventional methods focus on reducing either the input or output space, even though both are often reduced simultaneously in practice. Our coupled approach naturally supports goal-oriented dimension reduction, where either an input or output quantity of interest is prescribed. We consider, in particular, goal-oriented sensor placement and goal-oriented sensitivity analysis, which can be viewed as dimension reduction where the most important output or, respectively, input components are chosen. Both applications present difficult combinatorial optimization problems with expensive objectives such as the expected information gain and Sobol' indices. By optimizing gradient-based bounds, we can determine the most informative sensors and most influential parameters as the largest diagonal entries of some diagnostic matrices, thus bypassing the combinatorial optimization and objective evaluation.
Graph neural networks (GNNs) have emerged as a powerful tool for solving combinatorial optimization problems (COPs), exhibiting state-of-the-art performance in both graph-structured and non-graph-structured domains. However, existing approaches lack a unified framework capable of addressing a wide range of COPs. After presenting a summary of representative COPs and a brief review of recent advancements in GNNs for solving COPs, this paper proposes a unified framework for solving COPs based on GNNs, including graph representation of COPs, equivalent conversion of non-graph structured COPs to graph-structured COPs, graph decomposition, and graph simplification. The proposed framework leverages the ability of GNNs to effectively capture the relational information and extract features from the graph representation of COPs, offering a generic solution to COPs that can address the limitations of state-of-the-art in solving non-graph-structured and highly complex graph-structured COPs.
Constrained submodular optimization problems play a key role in the area of combinatorial optimization as they capture many NP-hard optimization problems. So far, Pareto optimization approaches using multi-objective formulations have been shown to be successful to tackle these problems while single-objective formulations lead to difficulties for algorithms such as the $(1+1)$-EA due to the presence of local optima. We introduce for the first time single-objective algorithms that are provably successful for different classes of constrained submodular maximization problems. Our algorithms are variants of the $(1+\lambda)$-EA and $(1+1)$-EA and increase the feasible region of the search space incrementally in order to deal with the considered submodular problems.
Job shop scheduling problems (JSSPs) represent a critical and challenging class of combinatorial optimization problems. Recent years have witnessed a rapid increase in the application of graph neural networks (GNNs) to solve JSSPs, albeit lacking a systematic survey of the relevant literature. This paper aims to thoroughly review prevailing GNN methods for different types of JSSPs and the closely related flow-shop scheduling problems (FSPs), especially those leveraging deep reinforcement learning (DRL). We begin by presenting the graph representations of various JSSPs, followed by an introduction to the most commonly used GNN architectures. We then review current GNN-based methods for each problem type, highlighting key technical elements such as graph representations, GNN architectures, GNN tasks, and training algorithms. Finally, we summarize and analyze the advantages and limitations of GNNs in solving JSSPs and provide potential future research opportunities. We hope this survey can motivate and inspire innovative approaches for more powerful GNN-based approaches in tackling JSSPs and other scheduling problems.
Recently, there has been much work on the design of general heuristics for graph-based, combinatorial optimization problems via the incorporation of Graph Neural Networks (GNNs) to learn distribution-specific solution structures.However, there is a lack of consistency in the evaluation of these heuristics, in terms of the baselines and instances chosen, which makes it difficult to assess the relative performance of the algorithms. In this paper, we propose an open-source benchmark suite MaxCut-Bench dedicated to the NP-hard Maximum Cut problem in both its weighted and unweighted variants, based on a careful selection of instances curated from diverse graph datasets. The suite offers a unified interface to various heuristics, both traditional and machine learning-based. Next, we use the benchmark in an attempt to systematically corroborate or reproduce the results of several, popular learning-based approaches, including S2V-DQN [31], ECO-DQN [4], among others, in terms of three dimensions: objective value, generalization, and scalability. Our empirical results show that several of the learned heuristics fail to outperform a naive greedy algorithm, and that only one of them consistently outperforms Tabu Search, a simple, general heuristic based upon local search. Furthermore, we find that the performance of ECO-DQN remains the same or is improved if the GNN is replaced by a simple linear regression on a subset of the features that are related to Tabu Search. Code, data, and pretrained models are available at: \url{https://github.com/ankurnath/MaxCut-Bench}.
In recent years, the use of data-driven methods has provided insights into underlying patterns and principles behind culinary recipes. In this exploratory work, we introduce the use of topological data analysis, especially persistent homology, in order to study the space of culinary recipes. In particular, persistent homology analysis provides a set of recipes surrounding the multiscale "holes" in the space of existing recipes. We then propose a method to generate novel ingredient combinations using combinatorial optimization on this topological information. We made biscuits using the novel ingredient combinations, which were confirmed to be acceptable enough by a sensory evaluation study. Our findings indicate that topological data analysis has the potential for providing new tools and insights in the study of culinary recipes.
Causal learning tackles the computationally demanding task of estimating causal graphs. This paper introduces a new divide-and-conquer approach for causal graph learning, called DCILP. In the divide phase, the Markov blanket MB($X_i$) of each variable $X_i$ is identified, and causal learning subproblems associated with each MB($X_i$) are independently addressed in parallel. This approach benefits from a more favorable ratio between the number of data samples and the number of variables considered. In counterpart, it can be adversely affected by the presence of hidden confounders, as variables external to MB($X_i$) might influence those within it. The reconciliation of the local causal graphs generated during the divide phase is a challenging combinatorial optimization problem, especially in large-scale applications. The main novelty of DCILP is an original formulation of this reconciliation as an integer linear programming (ILP) problem, which can be delegated and efficiently handled by an ILP solver. Through experiments on medium to large scale graphs, and comparisons with state-of-the-art methods, DCILP demonstrates significant improvements in terms of computational complexity, while preserving the learning accuracy on real-world problem and suffering at most a slight loss of accuracy on synthetic problems.
Combinatorial optimization (CO) is one of the most fundamental mathematical models in real-world applications. Traditional CO solvers, such as Branch-and-Bound (B&B) solvers, heavily rely on expert-designed heuristics, which are reliable but require substantial manual tuning. Recent studies have leveraged deep learning (DL) models as an alternative to capture rich feature patterns for improved performance on GPU machines. Nonetheless, the drawbacks of high training and inference costs, as well as limited interpretability, severely hinder the adoption of DL methods in real-world applications. To address these challenges, we propose a novel deep symbolic optimization learning framework that combines their advantages. Specifically, we focus on the node selection module within B&B solvers -- namely, deep symbolic optimization for node selection (Dso4NS). With data-driven approaches, Dso4NS guides the search for mathematical expressions within the high-dimensional discrete symbolic space and then incorporates the highest-performing mathematical expressions into a solver. The data-driven model captures the rich feature information in the input data and generates symbolic expressions, while the expressions deployed in solvers enable fast inference with high interpretability. Experiments demonstrate the effectiveness of Dso4NS in learning high-quality expressions, outperforming existing approaches on a CPU machine. Encouragingly, the learned CPU-based policies consistently achieve performance comparable to state-of-the-art GPU-based approaches.
Recently various optimization problems, such as Mixed Integer Linear Programming Problems (MILPs), have undergone comprehensive investigation, leveraging the capabilities of machine learning. This work focuses on learning-based solutions for efficiently solving the Quadratic Assignment Problem (QAPs), which stands as a formidable challenge in combinatorial optimization. While many instances of simpler problems admit fully polynomial-time approximate solution (FPTAS), QAP is shown to be strongly NP-hard. Even finding a FPTAS for QAP is difficult, in the sense that the existence of a FPTAS implies $P = NP$. Current research on QAPs suffer from limited scale and computational inefficiency. To attack the aforementioned issues, we here propose the first solution of its kind for QAP in the learn-to-improve category. This work encodes facility and location nodes separately, instead of forming computationally intensive association graphs prevalent in current approaches. This design choice enables scalability to larger problem sizes. Furthermore, a \textbf{S}olution \textbf{AW}are \textbf{T}ransformer (SAWT) architecture integrates the incumbent solution matrix with the attention score to effectively capture higher-order information of the QAPs. Our model's effectiveness is validated through extensive experiments on self-generated QAP instances of varying sizes and the QAPLIB benchmark.
2D irregular packing is a classic combinatorial optimization problem with various applications, such as material utilization and texture atlas generation. This NP-hard problem requires efficient algorithms to optimize space utilization. Conventional numerical methods suffer from slow convergence and high computational cost. Existing learning-based methods, such as the score-based diffusion model, also have limitations, such as no rotation support, frequent collisions, and poor adaptability to arbitrary boundaries, and slow inferring. The difficulty of learning from teacher packing is to capture the complex geometric relationships among packing examples, which include the spatial (position, orientation) relationships of objects, their geometric features, and container boundary conditions. Representing these relationships in latent space is challenging. We propose GFPack++, an attention-based gradient field learning approach that addresses this challenge. It consists of two pivotal strategies: \emph{attention-based geometry encoding} for effective feature encoding and \emph{attention-based relation encoding} for learning complex relationships. We investigate the utilization distribution between the teacher and inference data and design a weighting function to prioritize tighter teacher data during training, enhancing learning effectiveness. Our diffusion model supports continuous rotation and outperforms existing methods on various datasets. We achieve higher space utilization over several widely used baselines, one-order faster than the previous diffusion-based method, and promising generalization for arbitrary boundaries. We plan to release our source code and datasets to support further research in this direction.
Games, including abstract board games, constitute a convenient ground to create, design, and improve new AI methods. In this field, Monte Carlo Tree Search is a popular algorithm family, aiming to build game trees and explore them efficiently. Combinatorial Optimization, on the other hand, aims to model and solve problems with an objective to optimize and constraints to satisfy, and is less common in Game AI. We believe however that both methods can be combined efficiently, by injecting Combinatorial Optimization into Monte Carlo Tree Search to help the tree search, leading to a novel combination of these two techniques. Tested on the board game boop., our method beats 96% of the time the Monte Carlo Tree Search algorithm baseline. We conducted an ablation study to isolate and analyze which injections and combinations of injections lead to such performances. Finally, we opposed our AI method against human players on the Board Game Arena platform, and reached a 373 ELO rating after 51 boop. games, with a 69% win rate and finishing ranked 56th worldwide on the platform over 5,316 boop. players.
Millimeter wave (mmWave) cell-free massive MIMO (CF mMIMO) is a promising solution for future wireless communications. However, its optimization is non-trivial due to the challenging channel characteristics. We show that mmWave CF mMIMO optimization is largely an assignment problem between access points (APs) and users due to the high path loss of mmWave channels, the limited output power of the amplifier, and the almost orthogonal channels between users given a large number of AP antennas. The combinatorial nature of the assignment problem, the requirement for scalability, and the distributed implementation of CF mMIMO make this problem difficult. In this work, we propose an unsupervised machine learning (ML) enabled solution. In particular, a graph neural network (GNN) customized for scalability and distributed implementation is introduced. Moreover, the customized GNN architecture is hierarchically permutation-equivariant (HPE), i.e., if the APs or users of an AP are permuted, the output assignment is automatically permuted in the same way. To address the combinatorial problem, we relax it to a continuous problem, and introduce an information entropy-inspired penalty term. The training objective is then formulated using the augmented Lagrangian method (ALM). The test results show that the realized sum-rate outperforms that of the generalized serial dictatorship (GSD) algorithm and is very close to the upper bound in a small network scenario, while the upper bound is impossible to obtain in a large network scenario.
Mixed Integer Linear Programming (MILP) is a fundamental tool for modeling combinatorial optimization problems. Recently, a growing body of research has used machine learning to accelerate MILP solving. Despite the increasing popularity of this approach, there is a lack of a common repository that provides distributions of similar MILP instances across different domains, at different hardness levels, with standardized test sets. In this paper, we introduce Distributional MIPLIB, a multi-domain library of problem distributions for advancing ML-guided MILP methods. We curate MILP distributions from existing work in this area as well as real-world problems that have not been used, and classify them into different hardness levels. It will facilitate research in this area by enabling comprehensive evaluation on diverse and realistic domains. We empirically illustrate the benefits of using Distributional MIPLIB as a research vehicle in two ways. We evaluate the performance of ML-guided variable branching on previously unused distributions to identify potential areas for improvement. Moreover, we propose to learn branching policies from a mix of distributions, demonstrating that mixed distributions achieve better performance compared to homogeneous distributions when there is limited data and generalize well to larger instances. The dataset is publicly available at https://sites.google.com/usc.edu/distributional-miplib/home.
Quantum Annealing (QA) holds great potential for solving combinatorial optimization problems efficiently. However, the effectiveness of QA algorithms heavily relies on the embedding of problem instances, represented as logical graphs, into the quantum unit processing (QPU) whose topology is in form of a limited connectivity graph, known as the minor embedding Problem. Existing methods for the minor embedding problem suffer from scalability issues when confronted with larger problem sizes. In this paper, we propose a novel approach utilizing Reinforcement Learning (RL) techniques to address the minor embedding problem, named CHARME. CHARME includes three key components: a Graph Neural Network (GNN) architecture for policy modeling, a state transition algorithm ensuring solution validity, and an order exploration strategy for effective training. Through comprehensive experiments on synthetic and real-world instances, we demonstrate that the efficiency of our proposed order exploration strategy as well as our proposed RL framework, CHARME. In details, CHARME yields superior solutions compared to fast embedding methods such as Minorminer and ATOM. Moreover, our method surpasses the OCT-based approach, known for its slower runtime but high-quality solutions, in several cases. In addition, our proposed exploration enhances the efficiency of the training of the CHARME framework by providing better solutions compared to the greedy strategy.