analiza aiqbench
This research introduces ReTA, a reinforcement learning framework that dynamically decides when and how to integrate external knowledge graphs into electronic health record models for longitudinal prediction. By treating knowledge augmentation as a budget-aware policy with distinct import strategies, the method improves diagnostic accuracy on sparse medical data while maintaining computational efficiency. The approach demonstrates significant gains across multiple clinical tasks, offering a scalable solution to mitigate data sparsity in healthcare AI without indiscriminately expanding graph topology.
Import What You Need: Learning When and How to Augment EHR Graphs with External Knowledge
arXiv:2609.01839v1 Announce Type: new Abstract: Longitudinal prediction from electronic health records (EHRs) is limited by the sparsity and irregularity in patient trajectories, and knowledge augmentation with external knowledge graphs (KGs) offers a promising way to alleviate these issues. However, most existing methods perform fixed, context-agnostic topology augmentation by adding the same KG nodes and edges regardless of a patient's evolving state. We propose ReTA, a Reinforcement learning-based dynamic Topology Augmentation framework that casts KG import as a per-visit, budget-aware policy. ReTA first constructs an offline refined pool of KG-grounded templates, then learns a policy to select one augment action per visit from three options: Soft Import, which enriches node features without modifying graph topology, Hard Import, which grafts a compact KG subgraph onto the visit graph to create message-passing shortcuts, and Skip, which leaves the visit unaugmented when the base encoder is already confident. To stabilize learning, ReTA employs a decoupled encoder that processes semantic and structural signals in separate channels and fuses them via adaptive gating. Experiments on MIMIC-III and MIMIC-IV across diagnosis prediction, mortality, and readmission show that ReTA consistently outperforms strong baselines while remaining efficient, transfers across datasets and knowledge graphs, and yields interpretable augmentation patterns. The robust gains under sparse supervision highlight the advantage of ReTA's dynamic decision to import knowledge, boosting accuracy while curbing costs.
Abstract:Longitudinal prediction from electronic health records (EHRs) is limited by the sparsity and irregularity in patient trajectories, and knowledge augmentation with external knowledge graphs (KGs) offers a promising way to alleviate these issues. However, most existing methods perform fixed, context-agnostic topology augmentation by adding the same KG nodes and edges regardless of a patient's evolving state. We propose ReTA, a Reinforcement learning-based dynamic Topology Augmentation framework that casts KG import as a per-visit, budget-aware policy. ReTA first constructs an offline refined pool of KG-grounded templates, then learns a policy to select one augment action per visit from three options: Soft Import, which enriches node features without modifying graph topology, Hard Import, which grafts a compact KG subgraph onto the visit graph to create message-passing shortcuts, and Skip, which leaves the visit unaugmented when the base encoder is already confident. To stabilize learning, ReTA employs a decoupled encoder that processes semantic and structural signals in separate channels and fuses them via adaptive gating. Experiments on MIMIC-III and MIMIC-IV across diagnosis prediction, mortality, and readmission show that ReTA consistently outperforms strong baselines while remaining efficient, transfers across datasets and knowledge graphs, and yields interpretable augmentation patterns. The robust gains under sparse supervision highlight the advantage of ReTA's dynamic decision to import knowledge, boosting accuracy while curbing costs.
Accepted to the Main Conference of EMNLP 2026. 20 pages, 13 figures, and 18 tables. Code: this https URL
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2609.01839 [cs.LG]
(or arXiv:2609.01839v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.01839 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 331 cuvinte
Subiecte conexe
Presentation: Platform Teams Enabling AI - MCP/Multi-Agentic Tools across Linkedin
InfoQ · 5 iun. 2026
Do Explanations Increase the Risk of Decision Logic Leakage? Explanation-Guided Stealing of Graph Models
arXiv · 3 iun. 2026
Dual-Primal Graph VAEs for Noisy Label Aggregation
arXiv · 8 oct. 2026
Graph-centric agentic intelligence
Amazon Science · 1 oct. 2026