In high school timetabling, some institutions adopt a university-like structure in which students follow different curricula and multiple sections are offered for each course, without pre-assigning students to specific sections. In such settings, student sectioning and timetabling are strongly interdependent. Timetabling is usually carried out with integer linear programming or metaheuristics, before assigning students to sections on the fixed schedule. However, several Canadian high schools construct their schedules according to predefined block patterns designed to ensure a balanced distribution of instructional time. While this structure eliminates many classical timetabling constraints, it introduces a specific challenge: courses occupying half-blocks must be paired to form full blocks. This allows students to have the rest of their courses on well balanced, full blocks, thus easing student mixing between sections for improved student sectioning. The pairing of half-block courses is a critical step, as it directly impacts the balance of students across sections. Based on this context, this paper introduces a constraint programming approach to generate a feasible half-block courses schedule leading to good student balancing. The model is designed to enhance computational efficiency, accommodate additional practical constraints, and reduce the need for manual adjustments by scheduling operators. The approach is evaluated on anonymized real-world data from Canadian high schools for the 2024-2025 academic year, provided by Dash Computer Solutions, a company responsible for the annual design of many high school schedules in Quebec, and is integrated with their existing student sectioning algorithm. Our results show that our approach provides solutions of comparable quality to a human-guided local search with a hand-crafted initial solution, while reducing the designing time from 10 hours to an hour in the best case.

Lagrangian decomposition (LD) is a relaxation method that provides a dual bound for constrained optimization problems by decomposing them into more manageable sub-problems. This bound can be used in branch-and-bound algorithms to prune the search space effectively.In brief, a vector of Lagrangian multipliers is associated with each sub-problem, and an iterative procedure (e.g., a sub-gradient optimization) adjusts these multipliers to find the tightest bound. Initially applied to integer programming, Lagrangian decomposition also had success in constraint programming due to its versatility and the fact that global constraints provide natural sub-problems. However, the non-linear and combinatorial nature of sub-problems in constraint programming makes it computationally intensive to optimize the Lagrangian multipliers with sub-gradient methods at each node of the tree search. This currently limits the practicality of LD as a general bounding mechanism for constraint programming. To address this challenge, we propose a self-supervised learning approach that leverages neural networks to generate multipliers directly, yielding tight bounds. This approach significantly reduces the number of sub-gradient optimization steps required, enhancing the pruning efficiency and reducing the execution time of constraint programming solvers. This contribution is one of the few that leverage learning to enhance bounding mechanisms on the dual side, a critical element in the design of combinatorial solvers. This work presents a generic method for learning valid dual bounds in constraint programming. We validate our approach on two challenging combinatorial problems: The multi-dimensional knapsack problem and the shift scheduling problem. The results show that our approach can solve more instances than the standard application of LD to constraint programming, reduce execution time by more than half, and has promising generalization ability through fine-tuning.

The integration of constraint programming (CP) together with machine learning (ML) has emerged as a promising direction for tackling complex decision-making and combinatorial optimization problems. While CP offers expressive modeling capabilities and formal guarantees, ML provides adaptive methods for learning from data and generalizing across instances. This survey presents a comprehensive overview of recent advances in combining CP and ML. We first show how ML has been used to improve the CP toolbox, both in modeling and in the efficiency of solving. Then, we examine how CP can support ML, particularly in providing structure, guarantees, and symbolic reasoning capabilities. Finally, we identify key open challenges inherent to such hybrid approaches and outline promising directions for future research. This survey provides a first conceptual and structured review of recent advancements in this emerging field, aiming to serve as a resource for practitioners and researchers in both the CP and ML communities. To keep the progress up to date, a curated list of references is hosted on an accompanying repository (https://github.com/corail-research/CPML-paper-list) and is open to community contributions.

The effective management and control of building energy systems are crucial for reducing the energy consumption peak loads, CO2 emissions, and ensuring the stability of the power grid, while maintaining optimal comfort levels within buildings. The difficulty to accommodate this trade-off is amplified by dynamic environmental conditions and the need for scalable solutions that can adapt across various building types and geographic locations. Acknowledging the importance of this problem, NeurIPS conference hosted since 2020 the CityLearn control challenge to foster the design of innovative solutions in building energy management. Participants were tasked with developing strategies that not only enhance energy efficiency but also prioritize sustainability and occupant comfort. This paper introduces the Community-based Hierarchical Energy Systems Coordination Algorithm (CHESCA), the winning approach of the 2023 edition. We rely on a hierarchical approach adaptable to an arbitrary number of buildings, first optimizing building-level metrics individually, and later refining these through a central community-level controller to improve grid-related metrics. Compared to the other high-ranked competitors, our approach demonstrated fast inference capabilities like learning-based methods, while offering a better interpretability and a superior generalization capabilities with minimal data requirements. This paper details our approach, supported by comprehensive experimental results and ablation studies.

The ability of large language models (LLMs) to mimic human-like intelligence has led to a surge in LLM-based autonomous agents. Though recent LLMs seem capable of planning and reasoning given user instructions, their effectiveness in applying these capabilities for autonomous task solving remains underexplored. This is especially true in enterprise settings, where automated agents hold the promise of a high impact. To fill this gap, we propose WorkArena++, a novel benchmark consisting of 682 tasks corresponding to realistic workflows routinely performed by knowledge workers. WorkArena++ is designed to evaluate the planning, problem-solving, logical/arithmetic reasoning, retrieval, and contextual understanding abilities of web agents. Our empirical studies across state-of-the-art LLMs and vision-language models (VLMs), as well as human workers, reveal several challenges for such models to serve as useful assistants in the workplace. In addition to the benchmark, we provide a mechanism to effortlessly generate thousands of ground-truth observation/action traces, which can be used for fine-tuning existing models. Overall, we expect this work to serve as a useful resource to help the community progress towards capable autonomous agents. The benchmark can be found at https://github.com/ServiceNow/WorkArena.

Constraint programming is known for being an efficient approach to solving combinatorial problems. Important design choices in a solver are the branching heuristics, designed to lead the search to the best solutions in a minimum amount of time. However, developing these heuristics is a time-consuming process that requires problem-specific expertise. This observation has motivated many efforts to use machine learning to automatically learn efficient heuristics without expert intervention. Although several generic variable-selection heuristics are available in the literature, the options for value-selection heuristics are more scarce. We propose to tackle this issue by introducing a generic learning procedure that can be used to obtain a value-selection heuristic inside a constraint programming solver. This has been achieved thanks to the combination of a deep Q-learning algorithm, a tailored reward signal, and a heterogeneous graph neural network. Experiments on graph coloring, maximum independent set, and maximum cut problems show that this framework competes with the well-known impact-based and activity-based search heuristics and can find solutions close to optimality without requiring a large number of backtracks.

Decision diagrams are an increasingly important tool in cutting-edge solvers for discrete optimization. However, the field of decision diagrams is relatively new, and is still incorporating the library of techniques that conventional solvers have had decades to build. We drew inspiration from the warm-start technique used in conventional solvers to address one of the major challenges faced by decision diagram based methods. Decision diagrams become more useful the wider they are allowed to be, but also become more costly to generate, especially with large numbers of variables. In the original version of this paper, we presented a method of peeling off a sub-graph of previously constructed diagrams and using it as the initial diagram for subsequent iterations that we call peel-and-bound. We tested the method on the sequence ordering problem, and our results indicate that our peel-and-bound scheme generates stronger bounds than a branch-and-bound scheme using the same propagators, and at significantly less computational cost. In this extended version of the paper, we also propose new methods for using relaxed decision diagrams to improve the solutions found using restricted decision diagrams, discuss the heuristic decisions involved with the parallelization of peel-and-bound, and discuss how peel-and-bound can be hyper-optimized for sequencing problems. Furthermore, we test the new methods on the sequence ordering problem and the traveling salesman problem with time-windows (TSPTW), and include an updated and generalized implementation of the algorithm capable of handling any discrete optimization problem. The new results show that peel-and-bound outperforms ddo (a decision diagram based branch-and-bound solver) on the TSPTW. We also close 15 open benchmark instances of the TSPTW.

Combinatorial optimization is a well-established area in operations research and computer science. Until recently, its methods have focused on solving problem instances in isolation, ignoring that they often stem from related data distributions in practice. However, recent years have seen a surge of interest in using machine learning, especially graph neural networks, as a key building block for combinatorial tasks, either directly as solvers or by enhancing exact solvers. The inductive bias of GNNs effectively encodes combinatorial and relational input due to their invariance to permutations and awareness of input sparsity. This paper presents a conceptual review of recent key advancements in this emerging field, aiming at optimization and machine learning researchers.

Decision diagrams are an increasingly important tool in cutting-edge solvers for discrete optimization. However, the field of decision diagrams is relatively new, and is still incorporating the library of techniques that conventional solvers have had decades to build. We drew inspiration from the warm-start technique used in conventional solvers to address one of the major challenges faced by decision diagram based methods. Decision diagrams become more useful the wider they are allowed to be, but also become more costly to generate, especially with large numbers of variables. We present a method of peeling off a sub-graph of previously constructed diagrams and using it as the initial diagram for subsequent iterations that we call peel-and-bound. We test the method on the sequence ordering problem, and our results indicate that our peel-and-bound scheme generates stronger bounds than a branch-and-bound scheme using the same propagators, and at significantly less computational cost.

Prescriptive analytics provides organizations with scalable solutions for large-scale, automated decision making. At the core of prescriptive analytics methodology is optimization, a field devoted to the study of algorithms that solve complex decision-making problems. Optimization algorithms rely heavily on generic methods for identifying tight bounds, which provide both solutions to problems and optimality guarantees. In the last decade, decision diagrams (DDs) have demonstrated significant advantages in obtaining bounds compared with the standard linear relaxation commonly used by commercial solvers. However, the quality of the bounds computed by DDs depends heavily on the variable ordering chosen for the construction. Besides, the problem of finding an ordering that optimizes a given metric is generally NP-hard. This paper studies how machine learning, specifically deep reinforcement learning (DRL), can be used to improve bounds provided by DDs, in particular through learning a good variable ordering. The introduced DRL models improve primal and dual bounds, even over standard linear programming relaxations, and are integrated in a full-fledged branch-and-bound algorithm. This paper, therefore, provides a novel mechanism for utilizing machine learning to tighten bounds, adding to recent research on using machine learning to obtain high-quality heuristic solutions and, for the first time, using machine learning to improve relaxation bounds through a generic bounding method. We apply the methods on a classic optimization problem, the maximum independent set, and demonstrate through computational testing that optimization bounds can be significantly improved through DRL. We provide the code to replicate the results obtained on the maximum independent set. Summary of Contribution: This paper studies the use of reinforcement learning to compute a variable ordering of decision diagram-based approximations for discrete optimization problems. This is among the first works to propose the use of machine learning to improve upon generic bounding methods for discrete optimization problems, thereby establishing a critical bridge between optimization and learning.

Combinatorial optimization has found applications in numerous fields, from aerospace to transportation planning and economics. The goal is to find an optimal solution among a finite set of possibilities. The well-known challenge one faces with combinatorial optimization is the state-space explosion problem: the number of possibilities grows exponentially with the problem size, which makes solving intractable for large problems. In the last years, deep reinforcement learning (DRL) has shown its promise for designing good heuristics dedicated to solve NP-hard combinatorial optimization problems. However, current approaches have an important shortcoming: they only provide an approximate solution with no systematic ways to improve it or to prove optimality. In another context, constraint programming (CP) is a generic tool to solve combinatorial optimization problems. Based on a complete search procedure, it will always find the optimal solution if we allow an execution time large enough. A critical design choice, that makes CP non-trivial to use in practice, is the branching decision, directing how the search space is explored. In this work, we propose a general and hybrid approach, based on DRL and CP, for solving combinatorial optimization problems. The core of our approach is based on a dynamic programming formulation, that acts as a bridge between both techniques. We experimentally show that our solver is efficient to solve three challenging problems: the traveling salesman problem with time windows, the 4-moments portfolio optimization problem, and the 0-1 knapsack problem. Results obtained show that the framework introduced outperforms the stand-alone RL and CP solutions, while being competitive with industrial solvers.

End-to-end training of neural network solvers for combinatorial optimization problems such as the Travelling Salesman Problem is intractable and inefficient beyond a few hundreds of nodes. While state-of-the-art Machine Learning approaches perform closely to classical solvers when trained on trivially small sizes, they are unable to generalize the learnt policy to larger instances of practical scales. Towards leveraging transfer learning to solve large-scale TSPs, this paper identifies inductive biases, model architectures and learning algorithms that promote generalization to instances larger than those seen in training. Our controlled experiments provide the first principled investigation into such zero-shot generalization, revealing that extrapolating beyond training data requires rethinking the neural combinatorial optimization pipeline, from network layers and learning paradigms to evaluation protocols.

Combinatorial optimization is a well-established area in operations research and computer science. Until recently, its methods have mostly focused on solving problem instances in isolation, ignoring the fact that they often stem from related data distributions in practice. However, recent years have seen a surge of interest in using machine learning, especially graph neural networks, as a key building block for combinatorial tasks, either directly as solvers or by enhancing the former. This paper presents a conceptual review of recent key advancements in this emerging field, aiming at researchers in both optimization and machine learning.

Finding tight bounds on the optimal solution is a critical element of practical solution methods for discrete optimization problems. In the last decade, decision diagrams (DDs) have brought a new perspective on obtaining upper and lower bounds that can be significantly better than classical bounding mechanisms, such as linear relaxations. It is well known that the quality of the bounds achieved through this flexible bounding method is highly reliant on the ordering of variables chosen for building the diagram, and finding an ordering that optimizes standard metrics is an NP-hard problem. In this paper, we propose an innovative and generic approach based on deep reinforcement learning for obtaining an ordering for tightening the bounds obtained with relaxed and restricted DDs. We apply the approach to both the Maximum Independent Set Problem and the Maximum Cut Problem. Experimental results on synthetic instances show that the deep reinforcement learning approach, by achieving tighter objective function bounds, generally outperforms ordering methods commonly used in the literature when the distribution of instances is known. To the best knowledge of the authors, this is the first paper to apply machine learning to directly improve relaxation bounds obtained by general-purpose bounding mechanisms for combinatorial optimization problems.