collaborators
Showing cs.LGShow all

6 papers · 1 filter

cs.LG2026

NEST: Nested Event Stream Transformer for Sequences of Multisets

Minghui Sun, Haoyu Gong, Xingyu You +3

Event stream data often exhibit hierarchical structure in which multiple events co-occur, resulting in a sequence of multisets (i.e., bags of events). In electronic health records…

cs.LG2026

Multimodal Training to Unimodal Deployment: Leveraging Unstructured Data During Training to Optimize Structured Data Only Deployment

Zigui Wang, Minghui Sun, Jiang Shu +3

Unstructured Electronic Health Record (EHR) data, such as clinical notes, contain clinical contextual observations that are not directly reflected in structured data fields. This a…

cs.LG2026

Interval-Based AUC (iAUC): Extending ROC Analysis to Uncertainty-Aware Classification

Yuqi Li, Matthew M. Engelhard

In high-stakes risk prediction, quantifying uncertainty through interval-valued predictions is essential for reliable decision-making. However, standard evaluation tools like the r…

cs.LG2025

Borrowing From the Future: Enhancing Early Risk Assessment through Contrastive Learning

Minghui Sun, Matthew M. Engelhard, Benjamin A. Goldstein

Risk assessments for a pediatric population are often conducted across multiple stages. For example, clinicians may evaluate risks prenatally, at birth, and during Well-Child visit…

cs.LG2025

FairPOT: Balancing AUC Performance and Fairness with Proportional Optimal Transport

Pengxi Liu, Yi Shen, Matthew M. Engelhard +4

Fairness metrics utilizing the area under the receiver operator characteristic curve (AUC) have gained increasing attention in high-stakes domains such as healthcare, finance, and…

cs.LG2025

CLEAR: Unlearning Spurious Style-Content Associations with Contrastive LEarning with Anti-contrastive Regularization

Minghui Sun, Benjamin A. Goldstein, Matthew M. Engelhard

Learning representations unaffected by superficial characteristics is important to ensure that shifts in these characteristics at test time do not compromise downstream prediction…