Paper addresses high-dimensional linear regression with missing data, proposing efficient and nearly unbiased estimators.
problem High-dimensional linear regression with blockwise missing covariates and partially observed responses.
method Proposes a computationally efficient estimator and nearly unbiased debiased estimators using blockwise imputation and estimating equations.
result Asymptotically valid confidence intervals and statistical tests constructed based on debiased estimators.
New method handles missing data using AI for efficient inference.
problem Parameter estimation and inference with blockwise missing data.
method Tractable solution using AI models and semiparametric theory.
result IBM(RAY) and IBM(Adaptive) estimators achieve efficiency gains.
MSB framework improves survival prediction in immunotherapy patients with missing data.
problem High dimensionality and blockwise missingness in multimodal clinical data.
method MSB is a late-fusion framework that independently models modality-specific features before aggregating predictions via cross-validated stacking.
result MSB outperformed baseline algorithms in predicting progression-free survival in lung cancer patients.
Imputation-Powered Inference improves subpopulation efficiency in missing data settings.
problem Complex missing data patterns challenge standard inference methods.
method Imputation-Powered Inference (IPI) combines blackbox imputation with bias correction.
result IPI provides valid and efficient M-estimation under MCAR blockwise missingness.
Blockwise bootstrap improves ASR performance testing for correlated data.
problem Testing reliability of WER improvements between ASR systems.
method Divide evaluation utterances into nonoverlapping blocks and resample these blocks.
result The variance estimator of absolute WER difference is consistent under mild conditions.
Stochastic methods with coordinate-wise adaptive stepsize (such as RMSprop and Adam) have been widely used in training deep neural networks. Despite their fast convergence, they can generalize worse than stochastic gradient descent. In this paper, by revisiting the design of Adagrad, we propose to split the network par…
Mixture-of-experts (MoE) models are a powerful paradigm for modeling of data arising from complex data generating processes (DGPs). In this article, we demonstrate how different MoE models can be constructed to approximate the underlying DGPs of arbitrary types of data. Due to the probabilistic nature of MoE models, we…
Graphical lasso models ASR utterance dependencies for consistent WER estimation.
problem Modeling dependent structure among ASR utterances for accurate significance analysis.
method Graphical lasso for dependency modeling, followed by blockwise bootstrap resampling.
result Statistically consistent variance estimator of WER under mild conditions.
Proposes a Bayesian approach for integrating multiple linked matrices.
problem Integrating multiple linked matrices for diverse applications.
method Empirical Bayes Linked Matrix Decomposition (EB-LMD).
result Efficient estimation algorithm with no tuning parameters.
Proposes BONMI for integrating noisy matrices from multi-source data.
problem Integrating noisy matrices from multi-source data with block-wise missingness.
method Exploits orthogonal Procrustes problem to align eigenspaces and completes missing blocks.
result Statistical rate for eigenspace of underlying matrix comparable to independently missing assumption.
BSFP method reveals latent patterns in multi-omic data for predicting lung function in HIV-associated OLD.
problem Limited understanding of multi-omic molecular phenomena and clinical outcomes in obstructive lung disease.
method Bayesian Simultaneous Factorization and Prediction (BSFP) method for multi-omic data, accommodating imputation and full posterior inference.
result BSFP reveals distinct clusters of patients with OLD and multi-omic patterns related to lung function decline.
In this paper we purpose a blockwise descent algorithm for group-penalized multiresponse regression. Using a quasi-newton framework we extend this to group-penalized multinomial regression. We give a publicly available implementation for these in R, and compare the speed of this algorithm to a competing algorithm --- w…
Develops a two-level monotonic multistage recommender system for better user-specific prediction.
problem Leveraging user-item-stage dependencies in a monotonic chain of events for enhanced prediction accuracy.
method A multistage recommender system with a two-level monotonic property, using a large-margin classifier based on a nonnegative additive latent factor model.
result The proposed method outperforms existing methods in simulations and an article sharing dataset.
Paper finds sharpness differences in transformer blocks accelerating LLM training.
problem Understanding and accelerating large language model pre-training.
method Uncovering sharpness disparity across transformer blocks and proposing Blockwise Learning Rate.
result Blockwise Learning Rate strategy accelerates LLM pre-training with lower loss and speedup.
Unified approach for learning quantum operations from measurements.
problem Accurate reconstruction of unknown quantum operations from noisy measurements.
method Matrix sensing techniques, randomized measurement design, blockwise measurement design, alternating least squares (ALS).
result The proposed method provides theoretical guarantees for the identifiability and recovery of low-rank superoperators in the presence of noise.
Gradient-based temporal difference (GTD) algorithms are widely used in off-policy learning scenarios. Among them, the two time-scale TD with gradient correction (TDC) algorithm has been shown to have superior performance. In contrast to previous studies that characterized the non-asymptotic convergence rate of TDC only…
New method for credible intervals of Covid19 reproduction number.
problem Lack of credibility intervals in existing estimates.
method Combines Langevin Monte Carlo with Proximal operators.
result Produces credible intervals for reproduction number estimates.
Novel tensor perturbation bounds for orthogonal iteration methods.
problem Developing robust bounds for tensor reconstruction and subspace estimation.
method Blockwise tensor perturbation bounds for high-order orthogonal iteration (HOOI).
result Upper bounds for singular subspace estimation converge linearly and tensor reconstruction error bound is characterized by a simple quantity.
A framework estimates multiple precision matrices with shared structures.
problem Estimating multiple precision matrices with shared structures.
method Penalized likelihood framework with iterative algorithm alternating between convex and clustering problems.
result The method outperforms competitors and performs similarly to methods using prior information.
In this paper we combine two important extensions of ordinary least squares regression: regularization and optimal scaling. Optimal scaling (sometimes also called optimal scoring) has originally been developed for categorical data, and the process finds quantifications for the categories that are optimal for the regres…
We introduce GAMSEL (Generalized Additive Model Selection), a penalized likelihood approach for fitting sparse generalized additive models in high dimension. Our method interpolates between null, linear and additive models by allowing the effect of each variable to be estimated as being either zero, linear, or a low-co…
New model captures time series dependence across and within blocks.
problem Complex multivariate time series dependence structures.
method Time series Gaussian chain graph models with directed and undirected edges.
result Consistent recovery of time series chain graph structure.
Deep autoregressive sequence-to-sequence models have demonstrated impressive performance across a wide variety of tasks in recent years. While common architecture classes such as recurrent, convolutional, and self-attention networks make different trade-offs between the amount of computation needed per layer and the le…
Study on missing data mechanisms and simple imputation methods in fairness of machine learning algorithms.
problem Impact of missing data mechanisms and simple imputation methods on fairness of machine learning algorithms.
method Three popular datasets for classification fairness were used. Missing values were generated using three missing data mechanisms. Various missing data handling techniques (listwise deletion, mean imputation, mode imputation, multiple imputation) were applied to the datasets. Fairness was assessed using classification algorithms (random forests).
result Missing data mechanism does not significantly impact fairness; listwise deletion gives highest fairness on average.
Missing data imputation can help improve the performance of prediction models in situations where missing data hide useful information. This paper compares methods for imputing missing categorical data for supervised classification tasks. We experiment on two machine learning benchmark datasets with missing categorical…
Communication overhead is a major bottleneck hampering the scalability of distributed machine learning systems. Recently, there has been a surge of interest in using gradient compression to improve the communication efficiency of distributed neural network training. Using 1-bit quantization, signSGD with majority vote …
Optimal multiscale learning of linear operators
problem Statistical and computational limits of learning bounded linear operators between Sobolev spaces
method Reformulate as an infinite-dimensional matrix regression problem with heterogeneous multiscale structure
result Establish minimax rates and construct a finite-resolution blockwise least-squares estimator attaining these rates
Develops a VAE model for datasets with missing data.
problem Applying VAEs to datasets with missing data.
method A novel latent variable model of a corruption process generating missing data, with a tractable ELBO.
result Improved marginal log-likelihood and better missing data imputation compared to existing approaches.
Autoencoder neural network is implemented to estimate the missing data. Genetic algorithm is implemented for network optimization and estimating the missing data. Missing data is treated as Missing At Random mechanism by implementing maximum likelihood algorithm. The network performance is determined by calculating the…
New algorithms tackle complex multi-block optimization problems in machine learning.
problem Non-convex multi-block bilevel optimization with hierarchical sampling challenges.
method Blockwise stochastic variance-reduced methods with parallel speedup.
result Achieves matching complexity to single-block problems with parallel speedup.
New algorithms handle missing data to improve fairness in machine learning.
problem Missing values in data can lead to unfair outcomes in machine learning models.
method Developed scalable and adaptive algorithms to handle missing values while preserving predictive information.
result Our adaptive algorithms consistently achieve higher fairness and accuracy than standard impute-then-classify methods.
Trinary decision tree improves handling of missing data in machine learning.
problem Improving accuracy in decision tree algorithms when dealing with missing data.
method Introduces Trinary decision tree, which does not assume missing values contain information about the response.
result Trinary decision tree outperforms other algorithms in Missing Completely at Random settings, especially when data is only missing out-of-sample.
DeepIFSAC uses attention mechanisms and contrastive learning to impute missing values in tabular data.
problem Missing values in tabular data, especially when high and not random.
method Row and column attention in a contrastive learning framework with CutMix data augmentation.
result Proposed method outperforms state-of-the-art methods for missing rates between 10% and 90% and various missing value types.
This paper compares imputation and direct parameter estimation methods for missing data in correlation matrix visualization.
problem Missing data challenges in estimating correlation coefficients for accurate visualization.
method Comparison of imputation and direct parameter estimation methods for handling missing data.
result Direct parameter estimation (DPER) outperforms imputation for accurate correlation matrix visualization.
This tutorial covers methods for handling missing data in SP and ML.
problem Dealing with missing data in signal processing and machine learning.
method Grouping strategies into three tasks: imputation, estimation, and prediction.
result Promising and future research directions are discussed.
IFGAN uses feature-specific GANs for missing value imputation.
problem Missing value imputation in data mining.
method Feature-specific Generative Adversarial Networks (GAN).
result IFGAN outperforms state-of-the-art algorithms in various missing conditions.
Missing data enhances privacy in differential privacy.
problem Privacy preservation in datasets with missing values.
method Formalized missing data as a privacy amplification mechanism within differential privacy.
result Incomplete data can yield privacy amplification for differentially private algorithms.
The paper explores how missing data problems are related to causal inference.
problem Missing data in experiments makes causal inference difficult.
method The paper reinterprets missing data as a form of causal inference by considering counterfactual variables.
result Identification assumptions in missing data can be encoded using graphical models of counterfactual and observed variables.
We consider the problem of estimating the inverse covariance matrix by maximizing the likelihood function with a penalty added to encourage the sparsity of the resulting matrix. We propose a new approach based on the split Bregman method to solve the regularized maximum likelihood estimation problem. We show that our m…
We present an automatic classification method for astronomical catalogs with missing data. We use Bayesian networks, a probabilistic graphical model, that allows us to perform inference to pre- dict missing values given observed data and dependency relationships between variables. To learn a Bayesian network from incom…
A novel sequence-to-sequence model predicts missing sensor data.
problem Missing sensor data in sequences.
method Formulated a novel sequence-to-sequence model using forward and backward RNNs.
result The model produces the lowest errors in 12% more cases than the current state-of-the-art.
MIM adds indicator variables to improve model performance on incomplete data.
problem Missing data in incomplete data sets.
method Missing Indicator Method (MIM) and Selective MIM (SMIM).
result MIM improves model performance for informative missing values and high-dimensional data.
ELMV uses ensemble learning to handle missing values in EHR data.
problem Significant missing values in EHR data cause bias and unreliable conclusions.
method ELMV constructs multiple subsets with lower missing rates and uses a support set for ensemble learning.
result ELMV outperforms conventional methods in critical feature identification and outcome prediction.
Robust variable selection for high-dimensional data with missing and measurement errors.
problem Missing data and measurement errors confound data distribution.
method Exponential loss function with inverse probability weighting and additive error models.
result The Atan punishment method improves robust variable selection.
Two methods use BART to model missing data in leaf photosynthetic trait data.
problem Handling missing data in multivariate outcomes with non-ignorable mechanisms.
method Bayesian Additive Regression Trees (BART) for joint modeling of data and missingness indicators.
result Both methods effectively recover various missingness mechanisms and outperform existing approaches.
Gaussian Processes improve missing value imputation in datasets.
problem Handling missing values in large datasets.
method Sparse Gaussian Processes combined with stochastic variational inference.
result MGP significantly outperforms other imputation methods.
Sparse regression such as the Lasso has achieved great success in handling high-dimensional data. However, one of the biggest practical problems is that high-dimensional data often contain large amounts of missing values. Convex Conditioned Lasso (CoCoLasso) has been proposed for dealing with high-dimensional data with…
A new method for imputing missing data using graphical models.
problem Missing data in graphs and its impact on analysis.
method MMG framework based on conditional independence and PAI principle.
result Valid and efficient method for imputing missing data.