Differentiating Through Dual Prices: End-to-End Policy Learning Under Capacity Constraints

• URL: http://arxiv.org/abs/2608.04669v1
• Date: Wed, 05 Aug 2026 10:28:21 GMT
• Title: Differentiating Through Dual Prices: End-to-End Policy Learning Under Capacity Constraints
• Abstract summary: We study how to learn such an assignment policy from logged observational data.<n>Standard pipeline is decision-blind: fit one outcome model per arm by regression.<n>We instead train outcome models end-to-end, differentiating an off-policy estimate through the dual prices.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Many social services assign scarce resources, such as housing assistance or hospital interventions, to people who arrive one at a time: each arrival must receive a decision immediately, and the long-run usage of every resource must stay within its capacity. We study how to learn such an assignment policy from logged observational data. The standard pipeline is decision-blind: fit one outcome model per arm by regression, price each capacitated resource from the fitted models, and assign each arrival the arm whose predicted outcome minus price is largest. We instead train the outcome models end-to-end, differentiating an off-policy estimate of the deployed policy's value through the dual prices themselves. We study two formulations: an exact nonconvex one, and a convex relaxation whose optimum always satisfies the capacity constraints in expectation and which is suboptimal by at most a term linear in the smoothing temperature and logarithmic in the number of arms. Every method is evaluated in a queueing simulation with resources replenished at their capacity rates. Across six datasets, the two end-to-end variants take the top slots on a deployment-adjusted value index at every delay cost, including zero; when capacities are binding, decision-blind baselines frequently violate them and incur much longer queueing delays. On the largest dataset, a hospital cohort of seventy thousand patients, end-to-end training also achieves significantly higher policy value, a margin that survives a capacity-matched neural baseline. Flexible decision-blind regression remains the stronger pure predictor where ground truth is measurable; end-to-end training is best suited to settings where resources are genuinely scarce and feasibility matters.

Related papers

• Chance-constrained selection of sequential intervention strategies from counterfactual estimates [7.461214229363999]
  We present a predict-then-optimize framework for cost-based decisions.<n>In the prediction step, any estimator returning an outcome value and a cost distribution supplies what the decision rule consumes.<n>In the optimization step, a chance-constrained selection over a finite candidate set the bounds that the cumulative cost exceeds the budget.
  arXiv  Detail & Related papers  (2026-08-13T13:12:05Z)
• Capacity Confounds and Coverage Guarantees in Adaptive Sub-model Federated Learning [5.586191108738563]
  Sub-model federated learning lets resource-constrained clients train width-reduced versions of a global model, but existing methods allocate capacity by device resources alone.<n>We ask whether that step is possible, using HAS-FL, an adaptive capacity-allocation framework, as a test case.
  arXiv  Detail & Related papers  (2026-08-07T12:26:03Z)
• Generative Frontier Planning for Adaptive Peer-Referral Recruitment under Covariate-Dependent Arrivals [28.082278231005997]
  Peer-referral recruitment systems are critical for studying and intervening on hidden populations affected by infectious diseases.<n>We propose emphGenerative Frontier Planning (GFP), a model-based planner that replaces per-step Monte-Carlo sampling with a deterministic backup.<n>GFP outperforms random, reinforcement-learning, and i.i.d. dynamic-programming baselines across four discount factors.
  arXiv  Detail & Related papers  (2026-06-06T22:31:07Z)
• ZIP-RC: Optimizing Test-Time Compute via Zero-Overhead Joint Reward-Cost Prediction [57.799425838564]
  We present ZIP-RC, an adaptive inference method that equips models with zero-overhead inference-time predictions of reward and cost.<n> ZIP-RC improves accuracy by up to 12% over majority voting at equal or lower average cost.
  arXiv  Detail & Related papers  (2025-12-01T09:44:31Z)
• MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources [113.33902847941941]
  Variance-Aware Sampling (VAS) is a data selection strategy guided by Variance Promotion Score (VPS)<n>We release large-scale, carefully curated resources containing 1.6M long CoT cold-start data and 15k RL QA pairs.<n> Experiments across mathematical reasoning benchmarks demonstrate the effectiveness of both the curated data and the proposed VAS.
  arXiv  Detail & Related papers  (2025-09-25T14:58:29Z)
• Generalized Linear Bandits: Almost Optimal Regret with One-Pass Update [70.38810219913593]
  We study the generalized linear bandit (GLB) problem, a contextual multi-armed bandit framework that extends the classical linear model by incorporating a non-linear link function.<n>GLBs are widely applicable to real-world scenarios, but their non-linear nature introduces significant challenges in achieving both computational and statistical efficiency.<n>We propose a jointly efficient algorithm that attains a nearly optimal regret bound with $mathcalO(1)$ time and space complexities per round.
  arXiv  Detail & Related papers  (2025-07-16T02:24:21Z)
• Optimal patient allocation for echocardiographic assessments [0.0]
  We estimate patient no-show probabilities and derive empirical distributions of arrival times and exam durations.<n>We develop a discrete-event simulation model using SimPy and integrate it with the open source Python library.<n>Considering a hospital configuration with a 1:6 ratio of fetal to non-fetal rooms and a 4:2 ratio of fetal to non-fetal sonographers, we show that on-the-fly allocation generally yields better performance.
  arXiv  Detail & Related papers  (2025-05-17T17:51:23Z)
• Decision from Suboptimal Classifiers: Excess Risk Pre- and Post-Calibration [52.70324949884702]
  We quantify the excess risk incurred using approximate posterior probabilities in batch binary decision-making.<n>We identify regimes where recalibration alone addresses most of the regret, and regimes where the regret is dominated by the grouping loss.<n>On NLP experiments, we show that these quantities identify when the expected gain of more advanced post-training is worth the operational cost.
  arXiv  Detail & Related papers  (2025-03-23T10:52:36Z)
• Learning Optimal and Fair Policies for Online Allocation of Scarce Societal Resources from Data Collected in Deployment [4.659762505466962]
  We use administrative data collected in deployment to design an online policy that maximizes expected outcomes while satisfying budget constraints.<n>We show that using our policies improves rates of exit from homelessness by 5.16% and that policies that are fair in either allocation or outcomes by race come at a very low price of fairness.
  arXiv  Detail & Related papers  (2023-11-23T01:40:41Z)
• Pessimistic Minimax Value Iteration: Provably Efficient Equilibrium Learning from Offline Datasets [101.5329678997916]
  We study episodic two-player zero-sum Markov games (MGs) in the offline setting. The goal is to find an approximate Nash equilibrium (NE) policy pair based on a dataset collected a priori.
  arXiv  Detail & Related papers  (2022-02-15T15:39:30Z)
• Estimation of Optimal Dynamic Treatment Assignment Rules under Policy Constraints [0.0]
  We study estimation of an optimal dynamic treatment regime that guides the optimal treatment assignment for each individual at each stage based on their history. The paper proposes two estimation methods: one solves the treatment assignment problem sequentially through backward induction, and the other solves the entire problem simultaneously across all stages.
  arXiv  Detail & Related papers  (2021-06-09T12:42:53Z)
• Universal Off-Policy Evaluation [64.02853483874334]
  We take the first steps towards a universal off-policy estimator (UnO) We use UnO for estimating and simultaneously bounding the mean, variance, quantiles/median, inter-quantile range, CVaR, and the entire cumulative distribution of returns.
  arXiv  Detail & Related papers  (2021-04-26T18:54:31Z)
• Coordinated Online Learning for Multi-Agent Systems with Coupled Constraints and Perturbed Utility Observations [91.02019381927236]
  We introduce a novel method to steer the agents toward a stable population state, fulfilling the given resource constraints. The proposed method is a decentralized resource pricing method based on the resource loads resulting from the augmentation of the game's Lagrangian.
  arXiv  Detail & Related papers  (2020-10-21T10:11:17Z)

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.