10 papers
Sequential Certification of Threshold Decisions in Rare-Event Risk Prediction
Hui-Mean Foo, Yuan-chin Ivan Chang
Rare clinical outcomes pose a difficulty deeper than ordinary class imbalance: a penalized logistic model can return finite, stable-looking coefficients before the data support a r…
Estimating the Conditional Forecast-Revision Scale in Sequential Models: Local-Smoothing Limits, Matched Models, and Cost--Accuracy Trade-offs
Hui-Mean Foo, Yuan-chin Ivan Chang
The \emph{conditional forecast-revision scale} $\It=\{\Var(\E[X_{t+1}\mid\F_t]\mid\F_{t-1})\}^{1/2}$ measures the history-specific size of the forecast update induced by observing…
Coupling Precipitation Forecasting and Early Warning with Reverse-Martingale Recurrent Neural Networks
Hui-Mean Foo, Yuan-chin Ivan Chang
Precipitation forecasts are judged by accuracy, but the decisions they support -- when to restrict water, when to warn of drought -- turn on noticing when a local regime is becomin…
Backward Coherence and Hidden-State Stability in Recurrent Neural Networks: A Quasi-Reverse-Martingale Theory
Yuan-chin Ivan Chang
Recurrent neural networks maintain a hidden state , but its probabilistic meaning is often unclear. We study hidden-state stability through \emph{backward coherence}: the exte…
Target-Oriented Statistical Compression: Sufficiency, Reverse Martingales, and Sequential Monitoring
Yuan-chin Ivan Chang
Statistical procedures rarely retain all features of the observed data. A sufficient statistic removes information irrelevant to a parameter; a maximum likelihood estimate compress…
Small-Area Precipitation Forecasting and Drought--Flood Early Warning with Reverse-Martingale Regularized Recurrent Networks
Foo Hui-Mean, Yuan-chin Ivan Chang
Small-area precipitation forecasts support real-time decisions for reservoir operation, irrigation planning, drought monitoring, and flash-flood response. Operational value depends…
ALMAB-DC: Active Learning, Multi-Armed Bandits, and Distributed Computing for Sequential Experimental Design and Black-Box Optimization
Foo Hui-Mean, Yuan-chin I Chang
Sequential experimental design under expensive, gradient-free objectives is a central challenge in computational statistics: evaluation budgets are tightly constrained and informat…
Efficient Data Reduction Via PCA-Guided Quantile Based Sampling
Foo Hui-Mean, Yuan-chin Ivan Chang
In large-scale statistical modeling, reducing data size through subsampling is essential for balancing computational efficiency and statistical accuracy. We propose a new method, P…
Integrating Multi-Armed Bandit, Active Learning, and Distributed Computing for Scalable Optimization
Foo Hui-Mean, Yuan-chin Ivan Chang
Modern optimization problems in scientific and engineering domains often rely on expensive black-box evaluations, such as those arising in physical simulations or deep learning pip…
PCA-Guided Quantile Sampling: Preserving Data Structure in Large-Scale Subsampling
Foo Hui-Mean, Yuan-chin Ivan Chang
We introduce Principal Component Analysis guided Quantile Sampling (PCA QS), a novel sampling framework designed to preserve both the statistical and geometric structure of large s…