Stat.ML Papers

EHR-MPC: Inference-Time Control for Sepsis Treatment with Generative Patient Digital Twins ift.tt/Jb5saOF

Influence Diagnostics in High-dimensional M-estimation: Precise Asymptotics ift.tt/WCShkyr

Spectrally Deconfounded Gradient Boosting ift.tt/nEM9FVi

Characterization of the basin of convexity for multi-snapshot spike deconvolution via variable projection ift.tt/UT9pZN2

Deep Gaussian Processes on Directed Acyclic Graphs ift.tt/TJiacuw

Adaptive Bayes exactly tracks information over intrinsic time ift.tt/e32ZdQb

A Statistical Test for the Benefits of Personalizing Interventions ift.tt/aA0RBU9

Nonconvex Composite Functional Constraints via First-Order Augmented Lagrangian Methods under Local Regularity ift.tt/faP149S

Stochastic Linear Bandits with Partially Observed Actions ift.tt/qYohNrB

Optimal Top-$k$ Identification from Pairwise Comparisons ift.tt/kleOEg3

Achieving Almost Exact Recovery in Almost Quadratic Time: Rank-Based Graph Matching via Local Tree Correlation Tests ift.tt/FBCQrAp

Solving Stochastic Fixed-Point Equations with High Probability ift.tt/BbkWXJy

Similarity search generalisation in contrastive learning with InfoNCE loss ift.tt/Q3eSkW8

comprisk: A scikit-learn-compatible Python toolkit for competing-risks survival analysis ift.tt/Q7dwt3z

Near-optimal node-private community estimation in polynomial-time ift.tt/7WEV3p2

Deep Learning for Dynamic Programming with Recursive Utility Using First-order Conditions ift.tt/OqQjxdS

Neural Collapse Is Forbidden: Information Floors in Language Models ift.tt/gafrAK3

Terminal Dimension Reduction for Time Series with Applications ift.tt/o2G95ej

Statistically Undetectable Backdoors in Deep Neural Networks ift.tt/9DQujYw

High-Dimensional Interpolators Can Be Fragile: Heavy Tails and High-Dimensional Large Deviations ift.tt/pIPg0ye