@StatMLPapers profile picture
Stat.ML Papers @StatMLPapers

Unofficial updates of statistical machine learning papers on arXiv

Joined June 2012
  • Tweets 71,406
  • Following 0
  • Followers 25,285
Photos and videos
Identifiability Without Gaussianity: Symbolic World Models and Near-Infinite Temporal Consistency ift.tt/W6MlhKJ

Identifiability Without Gaussianity: Symbolic World Models and...

Klindt, LeCun, and Balestriero (arXiv:2605.26379) proved that Joint-Embedding Predictive Architectures (JEPAs) achieve linear identifiability, the linear recovery of the world's true latent...

arxiv.org
3
22
1,543
Epistemic Uncertainty Is Not the Reducible Kind ift.tt/8BlFHsY

Epistemic Uncertainty Is Not the Reducible Kind

The standard taxonomy of predictive uncertainty defines epistemic uncertainty as the part removable by collecting more data, while the standard measure identifies it with a mutual-information...

arxiv.org
2
696
Prediction-Powered Causal Inference by Automatic Debiased Machine Learning and Semi-Supervised Riesz Regression ift.tt/8DETGbp

Prediction-Powered Causal Inference by Automatic Debiased Machine...

This study investigates semiparametric efficient estimation of causal and structural parameters in a semi-supervised setting. In our setting, unlabeled auxiliary regressors are available in...

arxiv.org
2
12
1,591
Robust State-Conditional Feature-Weighted Jump Models for Temporal Clustering ift.tt/xyLFUdn

Robust State-Conditional Feature-Weighted Jump Models for Temporal...

We propose a robust feature-weighted jump model for time-dependent clustering. A penalty is used to encourage smoothness of transitions over time, while robustness is achieved through the use of a...

arxiv.org
1
546
ProtoX-AD: Self-Explainable Time Series Anomaly Detection and Characterization ift.tt/foONBQ4

ProtoX-AD: Self-Explainable Time Series Anomaly Detection and...

Recent advances in time series anomaly detection (TSAD) have highlighted the effectiveness of self-supervised classification-based approaches. These methods apply transformations to normal...

arxiv.org
4
613
Simultaneous Latent Budget Trees for Stratified Classification ift.tt/4gAlv8c

Simultaneous Latent Budget Trees for Stratified Classification

In the era of Explainable Artificial Intelligence, there is a renewed focus on single trees for their ease of interpretation. This paper introduces Simultaneous Latent Budget Trees, a...

arxiv.org
1
9
910
Majority-of-Three is Optimal ift.tt/f73ZxjV

Majority-of-Three is Optimal

We give a short proof that the majority vote of three independent consistent classifiers is an optimal learner in the realizable PAC setting. This proves optimality for the simplest voting scheme,...

arxiv.org
6
47
5,523
A Two-Parameter Weibull Framework for Diagnosing Transformer Weight Distributions ift.tt/RN7dyMi

A Two-Parameter Weibull Framework for Diagnosing Transformer...

We apply the Weibull distribution -- a two-parameter family from extreme-value theory -- as a diagnostic framework for element-wise weight magnitude distributions in transformers. At...

arxiv.org
3
858
Computationally tractable robust differentially private mean estimation ift.tt/GLJV9ql

Computationally tractable robust differentially private mean estimation

We develop a new, differentially private mean estimator called the balloon mean. The main features of the balloon mean are that it is computationally tractable and enjoys robustness to outlying...

arxiv.org
2
869
Physics-Informed Neural Networks for Chemotherapy Pharmacokinetics: Benchmarking the Clinical Estimator and Exposing Parameter Identifiability ift.tt/jL4ohfJ

Physics-Informed Neural Networks for Chemotherapy...

Physics-Informed Neural Networks (PINNs) are an attractive tool for partial-observation problems in biology, where the governing dynamics are known but some compartments cannot be measured....

arxiv.org
1
400
How Useful is Causal Invariance for Domain Adaptation in Finite-Sample Settings? ift.tt/09WulOw

How Useful is Causal Invariance for Domain Adaptation in...

Machine learning models often degrade when they are deployed on a target distribution that differs from the source distributions they were trained on. Recent work in causality-based domain...

arxiv.org
9
1,079
Two-Layer Linear Auto-Regressive Models Estimate Latent States ift.tt/xHumGbK

Two-Layer Linear Auto-Regressive Models Estimate Latent States

Auto-regressive models have emerged as powerful tools for sequential data, from language to video. Understanding how and why these models learn latent representations remains an open theoretical...

arxiv.org
1
4
673
A unified complexity bound for logconcave sampling ift.tt/D4ENg2b

A unified complexity bound for logconcave sampling

We give a simple, unified, and nearly tight bound for sampling arbitrary logconcave distributions from a warm start using the In-and-Out algorithm along with exponential lifting. The main new...

arxiv.org
2
937
On McDiarmid's Inequality under Dependence via Approximate Tensorization of Entropy ift.tt/maQwHuT

On McDiarmid's Inequality under Dependence via Approximate...

We argue that dependent versions of McDiarmid's inequality are a useful but underutilized tool in mathematical statistics, learning theory and theoretical computer science. To make this point, we...

arxiv.org
6
911
Diffusion-Network Alignment: An Efficient Algorithm and Explicit Probability Bounds ift.tt/vEQgpBz

Diffusion-Network Alignment: An Efficient Algorithm and Explicit...

This paper studies a variation of the classic network alignment problem, named diffusion-network alignment. The goal is to align the vertices of a rooted diffusion tree to the vertices of a...

arxiv.org
2
597
Reliability of Probabilistic Emulation of Physical Systems ift.tt/CBEfDLO

Reliability of Probabilistic Emulation of Physical Systems

Two dominant approaches have emerged for generating probabilistic forecasts of physical systems: generative models, such as diffusion or flow matching; and ensembles of deterministic models with...

arxiv.org
1
3
568
A Quadratic Order Reduction -- Gaussian Process Ordinary Differential Equation framework for the inference of Large Continuous Dynamical Systems ift.tt/JojDL4e

A Quadratic Order Reduction -- Gaussian Process Ordinary...

Forecasting the evolution of complex dynamical systems remains a fundamentally challenging task, primarily due to pronounced nonlinear interactions, high-dimensional state spaces, and the...

arxiv.org
3
15
890
Calibrating simplified vine copulas with a noise contrastive estimation approach ift.tt/s5cm09P

Calibrating simplified vine copulas with a noise contrastive...

Vine copulas provide a flexible framework for modeling complex multivariate dependence structures using only bivariate building blocks. Their practical success relies heavily on the simplifying...

arxiv.org
1
7
644
Towards More General Control of Diffusion Models Using Jeffrey Guidance ift.tt/uSkoWXe

Towards More General Control of Diffusion Models Using Jeffrey Guidance

A key strength of diffusion models lies in their flexibility, since their outputs can be controlled at sampling time through guidance. However, beyond simple cases such as conditional sampling,...

arxiv.org
4
22
1,548
REMAL: Residual Equilibrium Manifold Active Learning for Surrogate-Based Multidisciplinary Design Analysis ift.tt/ZqaxlgE

REMAL: Residual Equilibrium Manifold Active Learning for...

Multidisciplinary design analysis of coupled engineering systems requires the computation of equilibrium states in which all disciplinary coupling variables are mutually consistent. Conventional...

arxiv.org
1
8
838
Load more