NOCTA: Non-Greedy Objective Cost-Tradeoff Acquisition for Longitudinal Data

2025-07-16

AI Summary

To address the challenge of temporally adaptive feature acquisition for longitudinal data in resource-constrained critical domains (e.g., healthcare), this paper proposes NOCTA, a non-greedy objective-cost trade-off framework. NOCTA dynamically selects features during inference based on high information gain and low acquisition cost—measured in time, monetary expense, or patient risk—thereby enabling cost-aware sequential decision-making. It introduces the first unified estimation objective specifically designed for temporal feature acquisition and employs a dual-estimator architecture: NOCTA-NP estimates predictive uncertainty via a nearest-neighbor nonparametric strategy, while NOCTA-P directly models the predictive utility of feature acquisition. Evaluated on both synthetic and real-world clinical datasets, NOCTA significantly outperforms existing baselines, achieving comparable or superior prediction accuracy while reducing average feature acquisition cost by 23.6%–41.2%. This work establishes a new paradigm for low-cost, high-reliability longitudinal decision support under stringent resource constraints.

Technology Category

• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Machine Learning: Feature Construction/Reformulation
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

In many critical applications, resource constraints limit the amount of information that can be gathered to make predictions. For example, in healthcare, patient data often spans diverse features ranging from lab tests to imaging studies. Each feature may carry different information and must be acquired at a respective cost of time, money, or risk to the patient. Moreover, temporal prediction tasks, where both instance features and labels evolve over time, introduce additional complexity in deciding when or what information is important. In this work, we propose NOCTA, a Non-Greedy Objective Cost-Tradeoff Acquisition method that sequentially acquires the most informative features at inference time while accounting for both temporal dynamics and acquisition cost. We first introduce a cohesive estimation target for our NOCTA setting, and then develop two complementary estimators: 1) a non-parametric method based on nearest neighbors to guide the acquisition (NOCTA-NP), and 2) a parametric method that directly predicts the utility of potential acquisitions (NOCTA-P). Experiments on synthetic and real-world medical datasets demonstrate that both NOCTA variants outperform existing baselines.

Problem

• Optimizing feature acquisition under resource constraints for predictions
• Balancing information gain and cost in temporal data settings
• Improving longitudinal prediction via dynamic feature selection

Innovation

• Non-greedy feature acquisition considering cost tradeoffs
• Two estimators: non-parametric and parametric methods
• Optimizes temporal data collection for medical predictions