Decisions and Deployment: The Five-Year SAHELI Project (2020-2025) on Restless Multi-Armed Bandits for Improving Maternal and Child Health

• URL: http://arxiv.org/abs/2604.07384v1
• Date: Wed, 08 Apr 2026 03:22:26 GMT
• Title: Decisions and Deployment: The Five-Year SAHELI Project (2020-2025) on Restless Multi-Armed Bandits for Improving Maternal and Child Health
• Abstract summary: The SAHELI project ( 2020-2025) uses AI to allocate scarce resources in a maternal and child health program in India.<n>The SAHELI system solves this sequential resource allocation problem using a Restless Multi-Armed Bandit (RMAB) framework.<n>A key methodological innovation is the transition from a traditional Two-Stage "predict-then-optimize" approach to Decision-Focused Learning.
• Score: 30.733903110193264
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Maternal and child health is a critical concern around the world. In many global health programs disseminating preventive care and health information, limited healthcare worker resources prevent continuous, personalised engagement with vulnerable beneficiaries. In such scenarios, it becomes crucial to optimally schedule limited live-service resources to maximise long-term engagement. To address this fundamental challenge, the multi-year SAHELI project (2020-2025), in collaboration with partner NGO ARMMAN, leverages AI to allocate scarce resources in a maternal and child health program in India. The SAHELI system solves this sequential resource allocation problem using a Restless Multi-Armed Bandit (RMAB) framework. A key methodological innovation is the transition from a traditional Two-Stage "predict-then-optimize" approach to Decision-Focused Learning (DFL), which directly aligns the framework's learning method with the ultimate goal of maximizing beneficiary engagement. Empirical evaluation through large-scale randomized controlled trials demonstrates that the DFL policy reduced cumulative engagement drops by 31% relative to the current standard of care, significantly outperforming the Two-Stage model. Crucially, the studies also confirmed that this increased program engagement translates directly into statistically significant improvements in real-world health behaviors, notably the continued consumption of vital iron and calcium supplements by new mothers. Ultimately, the SAHELI project provides a scalable blueprint for applying sequential decision-making AI to optimize resource allocation in health programs.

Related papers

• Improving Access to Essential Medicines via Decision-Aware Machine Learning [19.04910713770059]
  We propose a novel decision-aware machine learning framework for essential medicines allocation.<n>In collaboration with the Sierra Leone national government, we performed a staggered, nationwide deployment of our system as a decision support tool.<n>Our work demonstrates how machine learning methods can improve efficiency at very low cost in resource-constrained global health settings.
  arXiv  Detail & Related papers  (2026-07-10T21:31:00Z)
• Information-Theoretic Reward Modeling for Stable RLHF: Detecting and Mitigating Reward Hacking [78.69179041551014]
  We propose an information-theoretic reward modeling framework based on the Information Bottleneck principle.<n>We show that InfoRM filters out preference-irrelevant information to alleviate reward misgeneralization.<n>We also introduce IBL, a distribution-level regularization that penalizes such deviations, effectively expanding the optimization landscape.
  arXiv  Detail & Related papers  (2025-10-15T15:51:59Z)
• Beyond Reactive Safety: Risk-Aware LLM Alignment via Long-Horizon Simulation [69.63626052852153]
  We propose a proof-of-concept framework that projects how model-generated advice could propagate through societal systems.<n>We also introduce a dataset of 100 indirect harm scenarios, testing models' ability to foresee adverse, non-obvious outcomes from seemingly harmless user prompts.
  arXiv  Detail & Related papers  (2025-06-26T02:28:58Z)
• IRL for Restless Multi-Armed Bandits with Applications in Maternal and Child Health [52.79219652923714]
  This paper is the first to present the use of inverse reinforcement learning (IRL) to learn desired rewards for RMABs.<n>We demonstrate improved outcomes in a maternal and child health telehealth program.
  arXiv  Detail & Related papers  (2024-12-11T15:28:04Z)
• Optimizing Vital Sign Monitoring in Resource-Constrained Maternal Care: An RL-Based Restless Bandit Approach [31.228987526386558]
  Wireless vital sign monitoring devices offer a labor-efficient solution for continuous monitoring. We devise an allocation algorithm for this problem by modeling it as a variant of the popular Restless Multi-Armed Bandit paradigm. We demonstrate in simulations that our approach outperforms the best baseline by up to a factor of $4$.
  arXiv  Detail & Related papers  (2024-10-10T21:20:07Z)
• Efficient Public Health Intervention Planning Using Decomposition-Based Decision-Focused Learning [33.14258196945301]
  We show how to exploit the structure of Restless Multi-Armed Bandits (RMABs) to speed up intervention planning. We use real-world data from an Indian NGO, ARMMAN, to show that our approach is up to two orders of magnitude faster than the state-of-the-art approach.
  arXiv  Detail & Related papers  (2024-03-08T21:31:00Z)
• A Decision-Language Model (DLM) for Dynamic Restless Multi-Armed Bandit Tasks in Public Health [29.894488663882328]
  Large Language Models (LLMs) have emerged as adept automated planners across domains of robotic control and navigation. We propose a Decision Language Model (DLM) for RMABs, enabling dynamic fine-tuning of RMAB policies using human-language commands.
  arXiv  Detail & Related papers  (2024-02-22T18:58:27Z)
• Deep Reinforcement Learning for Efficient and Fair Allocation of Health Care Resources [47.57108369791273]
  Scarcity of health care resources could result in the unavoidable consequence of rationing. There is no universally accepted standard for health care resource allocation protocols. We propose a transformer-based deep Q-network to integrate the disease progression of individual patients and the interaction effects among patients.
  arXiv  Detail & Related papers  (2023-09-15T17:28:06Z)
• Pruning the Way to Reliable Policies: A Multi-Objective Deep Q-Learning Approach to Critical Care [46.2482873419289]
  We introduce a deep Q-learning approach to obtain more reliable critical care policies. We evaluate our method in off-policy and offline settings using simulated environments and real health records from intensive care units.
  arXiv  Detail & Related papers  (2023-06-13T18:02:57Z)
• Field Study in Deploying Restless Multi-Armed Bandits: Assisting Non-Profits in Improving Maternal and Child Health [28.43878945119807]
  Cell phones have enabled non-profits to deliver critical health information to their beneficiaries in a timely manner. A key challenge in such information delivery programs is that a significant fraction of beneficiaries drop out of the program. We developed a Restless Multi-Armed Bandits system to help non-profits place crucial service calls for live interaction with beneficiaries to prevent such engagement drops.
  arXiv  Detail & Related papers  (2021-09-16T16:04:48Z)
• Contingency-Aware Influence Maximization: A Reinforcement Learning Approach [52.109536198330126]
  influence (IM) problem aims at finding a subset of seed nodes in a social network that maximize the spread of influence. In this study, we focus on a sub-class of IM problems, where whether the nodes are willing to be the seeds when being invited is uncertain, called contingency-aware IM. Despite the initial success, a major practical obstacle in promoting the solutions to more communities is the tremendous runtime of the greedy algorithms.
  arXiv  Detail & Related papers  (2021-06-13T16:42:22Z)

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.