A new tree-based density kernel with logit-normal splitting probabilities is developed for nonparametric hierarchical mixture models and applied to cluster DNase-seq profiles from ENCODE, yielding clusters aligned with TF binding.
" * write output.state after.block = add.period write newline
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
A fiducial approach is developed for semiparametric Cox models that performs well when the MLE fails.
A Dirichlet process mixture model of GEV distributions for heterogeneous block maxima in extreme value analysis.
Develops a locally optimal design criterion for A/B testing on undirected networks under a CAR model that accounts for treatment, covariates, and network-induced correlations, with robustness checks via examples.
Proposes Mm-Φ_p design maximizing minimum Φ_p-efficiency under model uncertainties for multivariate GLM, along with a convergent algorithm for its construction.
MAS augments subsampling-based MLE with full-data moments through GMM to obtain smaller asymptotic variance and potentially full-data efficiency.
Active learning framework that combines D-optimality with maximin space-filling via Gaussian process surrogates to recover governing differential equations with fewer experiments than standard designs.
A statistical survey of RLHF for LLM alignment that connects preference learning and policy optimization to models like Bradley-Terry-Luce while reviewing methods, extensions, and open challenges.
citing papers explorer
-
A tree-based kernel for densities and its applications in clustering DNase-seq profiles
A new tree-based density kernel with logit-normal splitting probabilities is developed for nonparametric hierarchical mixture models and applied to cluster DNase-seq profiles from ENCODE, yielding clusters aligned with TF binding.
-
Semiparametric fiducial inference for Cox models
A fiducial approach is developed for semiparametric Cox models that performs well when the MLE fails.
-
Bayesian Mixture Models for Heterogeneous Extremes
A Dirichlet process mixture model of GEV distributions for heterogeneous block maxima in extreme value analysis.
-
Locally Optimal Design for A/B Testing in the Presence of Covariates and Network Connection
Develops a locally optimal design criterion for A/B testing on undirected networks under a CAR model that accounts for treatment, covariates, and network-induced correlations, with robustness checks via examples.
-
A Maximin $\Phi_{p}$-Efficient Design for Multivariate GLM
Proposes Mm-Φ_p design maximizing minimum Φ_p-efficiency under model uncertainties for multivariate GLM, along with a convergent algorithm for its construction.
-
A Moment-assisted Approach for Improving Subsampling-based MLE with Large-scale data
MAS augments subsampling-based MLE with full-data moments through GMM to obtain smaller asymptotic variance and potentially full-data efficiency.
-
Gaussian Process Assisted Active Learning of Physical Laws
Active learning framework that combines D-optimality with maximin space-filling via Gaussian process surrogates to recover governing differential equations with fewer experiments than standard designs.
-
Reinforcement Learning from Human Feedback: A Statistical Perspective
A statistical survey of RLHF for LLM alignment that connects preference learning and policy optimization to models like Bradley-Terry-Luce while reviewing methods, extensions, and open challenges.