Hard Constraints, Smooth Gradients: Learning Feasible Inventory Policies via Differentiable Projection

• URL: http://arxiv.org/abs/2608.02343v1
• Date: Mon, 03 Aug 2026 14:57:21 GMT
• Title: Hard Constraints, Smooth Gradients: Learning Feasible Inventory Policies via Differentiable Projection
• Abstract summary: Mixed-integer linear programs (MILPs) handle such constraints flexibly but scale poorly in environments.<n>We bridge this gap by embedding a differentiable convex optimization module inside the policy.<n>We show that our feasibility enforcement has bounded error relative to an exact integer projection and ensures the entire feasible action space is reachable.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Many operational problems are constrained sequential decision processes with large, combinatorial action spaces and interdependent feasibility constraints. Mixed-integer linear programs (MILPs) handle such constraints flexibly but scale poorly in stochastic environments. Deep reinforcement learning (DRL) promises scalable decision rules, but existing methods either penalize constraints rather than enforce them, or rely on feasibility mechanisms that break down once constraints interact. We bridge this gap by embedding a differentiable convex optimization module inside the policy: a neural network proposes continuous action targets, a quadratic program projects them onto the relaxed feasible set, and a dual-informed integer mapping restores integrality while preserving feasibility. Given a differentiable simulator, the policy trains end to end from sampled trajectories using pathwise gradients, while handling hard constraints with similar flexibility to MILPs. We show that our feasibility enforcement has bounded error relative to an exact integer projection and ensures the entire feasible action space is reachable. We apply the method to multi-echelon production-inventory planning under shared resource and material constraints. Our policy attains an average optimality gap below 1% on small instances. It further outperforms state-of-the-art echelon base-stock policies by up to 9.75% and a rolling-horizon multi-stage stochastic program by at least 7.7% in larger networks. On an industry-scale case study from ASML, it reduces average cost by up to 3.22% relative to the best-known benchmark policy. The savings are largest where planning is hardest: in tightly capacitated systems with high demand variability. More broadly, our work shows that DRL can deliver economically significant savings in sequential decision problems with interdependent hard constraints, which are widespread in practice.

Related papers

• Adaptive Neighborhood-Constrained Q Learning for Offline Reinforcement Learning [52.03884701766989]
  offline reinforcement learning (RL) algorithms typically impose constraints on action selection.<n>We propose a new neighborhood constraint that restricts action selection in the Bellman target to the union of neighborhoods of dataset actions.<n>We develop a simple yet effective algorithm, Adaptive Neighborhood-constrained Q learning (ANQ), to perform Q learning with target actions satisfying this constraint.
  arXiv  Detail & Related papers  (2025-11-04T13:42:05Z)
• Constrained Adversarial Perturbation [16.05659740749269]
  Universal Adversarial Perturbations (UAPs) have emerged as a powerful tool for both stress testing model robustness and scalable adversarial training.<n>We propose Constrained Adversarial Perturbation (CAP), an efficient algorithm that solves this problem using a gradient based alternating optimization strategy.
  arXiv  Detail & Related papers  (2025-10-17T14:44:20Z)
• Constrained Gaussian Process Motion Planning via Stein Variational Newton Inference [16.97959227638119]
  We propose a novel constrained Stein Variational Gaussian Process Motion Planning (cSGPMP) framework.<n>Our approach improves the efficiency of particle-based inference while explicitly handling nonlinear constraints.<n>We validate our method on standard benchmarks, achieving an average success rate of 98.57% across 350 planning tasks.
  arXiv  Detail & Related papers  (2025-04-07T11:20:11Z)
• Reinforcement learning with combinatorial actions for coupled restless bandits [62.89013331120493]
  We propose SEQUOIA, an RL algorithm that directly optimize for long-term reward over the feasible action space.<n>We empirically validate SEQUOIA on four novel restless bandit problems with constraints: multiple interventions, path constraints, bipartite matching, and capacity constraints.
  arXiv  Detail & Related papers  (2025-03-01T21:25:21Z)
• Efficiently Training Deep-Learning Parametric Policies using Lagrangian Duality [55.06411438416805]
  Constrained Markov Decision Processes (CMDPs) are critical in many high-stakes applications.<n>This paper introduces a novel approach, Two-Stage Deep Decision Rules (TS- DDR) to efficiently train parametric actor policies.<n>It is shown to enhance solution quality and to reduce computation times by several orders of magnitude when compared to current state-of-the-art methods.
  arXiv  Detail & Related papers  (2024-05-23T18:19:47Z)
• Learning Adversarial MDPs with Stochastic Hard Constraints [37.24692425018]
  We study online learning in constrained Markov decision processes (CMDPs) with adversarial losses and hard constraints.<n>Our work is the first to study CMDPs involving both adversarial losses and hard constraints.
  arXiv  Detail & Related papers  (2024-03-06T12:49:08Z)
• Uniformly Safe RL with Objective Suppression for Multi-Constraint Safety-Critical Applications [73.58451824894568]
  The widely adopted CMDP model constrains the risks in expectation, which makes room for dangerous behaviors in long-tail states. In safety-critical domains, such behaviors could lead to disastrous outcomes. We propose Objective Suppression, a novel method that adaptively suppresses the task reward maximizing objectives according to a safety critic.
  arXiv  Detail & Related papers  (2024-02-23T23:22:06Z)
• Generative Modelling of Stochastic Actions with Arbitrary Constraints in Reinforcement Learning [25.342811509665097]
  Many problems in Reinforcement Learning (RL) seek an optimal policy with large discrete multidimensional yet unordered action spaces. A challenge in this setting is that the underlying action space is categorical (discrete and unordered) and large. In this work, we address these challenges by applying a (state) conditional normalizing flow to compactly represent the policy.
  arXiv  Detail & Related papers  (2023-11-26T15:57:20Z)
• Action-Quantized Offline Reinforcement Learning for Robotic Skill Learning [68.16998247593209]
  offline reinforcement learning (RL) paradigm provides recipe to convert static behavior datasets into policies that can perform better than the policy that collected the data. In this paper, we propose an adaptive scheme for action quantization. We show that several state-of-the-art offline RL methods such as IQL, CQL, and BRAC improve in performance on benchmarks when combined with our proposed discretization scheme.
  arXiv  Detail & Related papers  (2023-10-18T06:07:10Z)
• Combining Deep Learning and Optimization for Security-Constrained Optimal Power Flow [94.24763814458686]
  Security-constrained optimal power flow (SCOPF) is fundamental in power systems. Modeling of APR within the SCOPF problem results in complex large-scale mixed-integer programs. This paper proposes a novel approach that combines deep learning and robust optimization techniques.
  arXiv  Detail & Related papers  (2020-07-14T12:38:21Z)
• Deep Constrained Q-learning [15.582910645906145]
  In many real world applications, reinforcement learning agents have to optimize multiple objectives while following certain rules or satisfying a set of constraints. We propose Constrained Q-learning, a novel off-policy reinforcement learning framework restricting the action space directly in the Q-update to learn the optimal Q-function for the induced constrained MDP and the corresponding safe policy.
  arXiv  Detail & Related papers  (2020-03-20T17:26:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.