ESRF reduces ARF ensemble size without sacrificing accuracy.
problem Over-provisioning of ARF ensemble leads to high CPU and memory consumption.
method ESRF uses a swap and elastic component to dynamically adjust the number of classifiers.
result ESRF reduces the number of classifiers by up to one third without sacrificing accuracy.
EM algorithm converges globally for two-component mixed linear regression.
problem Global convergence of EM algorithm for mixed linear regression.
method Developed new theoretical analysis for EM algorithm convergence in mixed linear regression.
result EM algorithm converges globally for two-component mixed linear regression.
EM algorithm converges linearly and achieves sharp rate in estimating mixtures of pairwise differences.
problem Estimating mixtures of pairwise differences from noisy data.
method Sharp analysis of the EM algorithm locally around the ground truth.
result The EM sequence converges linearly with an ℓ∞-norm guarantee on the estimation error and achieves the sharp rate of estimation in the ℓ2-norm. Proposes an EM method for learning from positive and unlabeled data with random selection assumption.
problem Learning from positive and unlabeled data with random selection assumption.
method Proposes an EM method to learn under the assumption that positive examples are selected at random, conditioned on some attributes.
result The proposed method outperforms state-of-the-art methods for learning under the selected completely at random assumption.
This paper analyzes EM algorithm for softmax mixture models in high dimensions.
problem Modeling heterogeneous populations choosing from multiple attributes.
method Comprehensive analysis of the EM algorithm for softmax mixture models (SMMs), proving identifiability and convergence.
result EM algorithm recovers mixture atoms at near-parametric rate under suitable initialization.
DP-EM algorithm improves privacy in EM iterations.
problem Privacy in iterative EM algorithms.
method Novel moment perturbation and composition methods (MA, zCDP).
result Reduces privacy noise in EM iterations.
Efficient inference for nonparametric Hawkes processes using Pólya-Gamma augmentation.
problem Efficient inference for nonparametric Hawkes processes.
method Pólya-Gamma augmentation, EM algorithm, mean-field variational inference.
result The proposed algorithms can recover well the underlying prompting characteristics efficiently.
New model tackles interference in online experiments.
problem Understanding cumulative performance in interference experiments.
method Introduces Multi-Armed Bandits with Interference (MABI) model.
result Cluster randomization policy achieves optimal expected regret and high probability bound.
LMLFM tackles predictive modeling from longitudinal data with mixed correlations.
problem Learning predictive models from longitudinal data with complex correlations and non-linear interactions.
method Longitudinal Multi-Level Factorization Machine (LMLFM) that selects predictive fixed and random effects.
result LMLFM outperforms state-of-the-art methods in predictive accuracy, variable selection, and scalability.
We develop a general framework for proving rigorous guarantees on the performance of the EM algorithm and a variant known as gradient EM. Our analysis is divided into two parts: a treatment of these algorithms at the population level (in the limit of infinite data), followed by results that apply to updates based on a …
Cryo-EM reconstruction is reformulated as a stochastic inverse problem to handle structural heterogeneity.
problem Handling structural heterogeneity in cryo-EM 3D reconstruction.
method Formulated as a stochastic inverse problem over probability measures, using variational discrepancy and Wasserstein gradient flow.
result Validated approach using synthetic examples, demonstrating recovery of continuous structural distributions.
Paper analyzes EM algorithm's trajectory in 2MLR, revealing cycloid behavior.
problem Understanding the convergence and trajectory of EM algorithm in 2MLR.
method Explicit closed-form expressions for EM updates, recurrence relation derivation at population level.
result EM iterations lie on a cycloid trajectory, leading to theoretical estimate of convergence exponent.
Machine learning models predict depression risk based on various factors.
problem Identifying individuals at greatest risk for depression.
method Random Effects/Expectation Maximization (RE-EM) trees and Mixed Effects Random Forest (MERF) algorithms.
result Machine learning models accurately predict depression severity and identify key predictors.
Strongly polynomial algorithm for approximate Forster transforms and halfspace learning.
problem Computing approximate Forster transforms and halfspace learning.
method Strongly polynomial time algorithm for approximate Forster transforms and halfspace learning.
result First strongly polynomial time algorithm for distribution-free PAC learning of halfspaces.
The paper simplifies complex tree ensembles for better interpretability.
problem Limited interpretability of tree ensembles like random forest and boosted trees.
method A post-processing method that approximates complex tree ensembles with a simpler, interpretable model using the EM algorithm.
result Complex tree ensembles can be approximated reasonably by simpler, interpretable models.
Simple randomized EM method outperforms state-of-the-art DA methods.
problem Classifying unlabeled target data using labeled source data from a related domain.
method Randomized Expectation Maximization (EM) method applied to logistic regression and support vector machine.
result Achieves state-of-the-art results on 36 real-life adaptation tasks.
Determining the 3D structures of biological molecules is a key problem for both biology and medicine. Electron Cryomicroscopy (Cryo-EM) is a promising technique for structure estimation which relies heavily on computational methods to reconstruct 3D structures from 2D images. This paper introduces the challenging Cryo-…
The Levy-Ito theorem explains how certain random processes can be broken down.
problem Understanding how certain random processes can be decomposed.
method Martingale methods are used to prove the Lévy-Ito decomposition theorem.
result The Lévy-Khintchine representation of infinitely divisible distributions is derived.
Paper introduces EM-KSH and EM-SPLH methods for supervised hashing.
problem Efficiently optimize retrieval speed and storage cost while preserving semantic information.
method Convert supervised hashing formulations to CRF, solve consistency equations using linear approximation of sigmoid function.
result Experimental results show superior performance of EM-KSH and EM-SPLH.
Two EM algorithms estimate prior distributions in mixture of linear regressions.
problem Estimating prior distributions in mixture of linear regressions.
method Two EM algorithms: one for continuous priors, one for discrete priors.
result Both algorithms accurately estimate prior distributions and the number of clusters.
EM algorithm achieves optimal sample complexity for learning two-component mixed linear regression.
problem Learning two-component mixed linear regression under varying signal-to-noise ratios.
method Analysis of EM algorithm convergence rates under different SNR regimes.
result EM algorithm achieves minimax optimal sample complexity in all SNR regimes.
EM algorithm improves accuracy in estimating Gaussian mixture centers.
problem Estimating centers of a two-component Gaussian mixture.
method Extending statistical analysis of EM algorithm using Stein's Lemma.
result Significantly expanded the basin of attraction for valid initialization.
We investigate the ergodic problem of growth-rate maximization under a class of risk constraints in the context of incomplete, Itô-process models of financial markets with random ergodic coefficients. Including {\em value-at-risk} (VaR), {\em tail-value-at-risk} (TVaR), and {\em limited expected loss} (LEL), these cons…
EM algorithm converges in 10 steps for 2 Gaussian mixtures.
problem Estimating mixtures of two Gaussians with known covariance matrices.
method Expectation-Maximization (EM) algorithm with convergence guarantees.
result EM converges geometrically in 10 steps for 2 Gaussian mixtures.
We examine methods for clustering in high dimensions. In the first part of the paper, we perform an experimental comparison between three batch clustering algorithms: the Expectation-Maximization (EM) algorithm, a winner take all version of the EM algorithm reminiscent of the K-means algorithm, and model-based hierarch…
EM algorithm converges to true mean for truncated mixtures of Gaussians.
problem Analyzing EM algorithm for truncated mixtures of two Gaussians.
method Using dynamical systems, probability, and statistics techniques.
result EM converges almost surely to true mean for various measurable sets S.
A new distortion measure optimizes function approximations in vector quantization.
problem Measuring the quality of vector quantization points for natural signals.
method A canonical distortion measure (CDM) is introduced, induced by an environment of functions on input space.
result Optimizing reconstruction error with respect to CDM yields optimal piecewise constant approximations.
New EM algorithms for weighted-data clustering improve audio-visual scene analysis.
problem Improving clustering of weighted data in heterogeneous environments.
method Proposed weighted-data Gaussian mixture model and two EM algorithms.
result Validation shows improved clustering in audio-visual scenes.
Efficiently clusters incomplete data without imputation or full EM, faster and more accurate.
problem Clustering partially recorded data efficiently.
method Model-based approach using multivariate t-distributions, considering only observed values.
result Approach is more accurate and computationally efficient than alternatives.
Proposes a robust EM algorithm for analyzing incomplete panel count data.
problem Missing reports in panel count data.
method Functional EM algorithm for non-parametric counting process mean function estimation.
result Robust to misspecification of Poisson process assumption and missing completely at random.
New numerical method for non-linear asset price model with CEV volatility.
problem Describing stochastic volatility in asset price dynamics.
method Proposes a mean-reverting theta-rho model with CEV volatility, constructs a truncated EM method.
result Truncated EM solutions can evaluate path-dependent financial products.
Study reveals bad local maxima in Gaussian mixture models, affecting EM algorithm performance.
problem Bad local maxima in Gaussian mixture models' likelihood function.
method Analyzes population likelihood function and EM algorithm convergence.
result EM algorithm can converge to bad local maxima with high probability.
New algorithms update dynamic graph regression faster than existing methods.
problem Efficiently updating linear regression solutions for dynamic graphs.
method Subsampled randomized Hadamard transform and CountSketch.
result First sublinear update time randomized algorithms for dynamic graph regression.
Paper proposes efficient and accurate initialization and EM algorithm for PL mixture models.
problem Initialization issues and combinatorial complexity in PL likelihood maximization.
method Initialization algorithm and EM algorithm for true log-likelihood maximization.
result Proposed algorithm provides accurate initial estimates and efficiently maximizes true log-likelihood.
GMNN combines conditional random fields and graph neural networks for relational data.
problem Semi-supervised object classification in relational data.
method Combines conditional random fields and graph neural networks. Uses variational EM algorithm for training.
result GMNN achieves state-of-the-art results on object classification, link classification, and unsupervised node representation learning.
In this paper we propose new techniques to sample arbitrary third-order tensors, with an objective of speeding up tensor algorithms that have recently gained popularity in machine learning. Our main contribution is a new way to select, in a biased random way, only O(n1.5/ε2) of the possible n3 elements while s…
New method reduces distributed non-convex optimization rounds and bits.
problem Efficiently optimizing non-convex models in distributed systems.
method Introduces permutation compressors to reduce communication complexity.
result PermK compressors lead to significant communication complexity improvements.
The study examines statistical inference with gradient ascent in multi-modal likelihood functions.
problem Statistical inference with multiple initializations in multi-modal likelihood functions.
method Derives population quantity, studies asymptotic normality, bootstrap, and likelihood ratio tests.
result Coverage deficiency and differences in CIs due to finite number of initializations.
We describe a method to perform functional operations on probability distributions of random variables. The method uses reproducing kernel Hilbert space representations of probability distributions, and it is applicable to all operations which can be applied to points drawn from the respective distributions. We refer t…
Combines VAMP with EM for joint signal and parameter recovery.
problem Recovering signals from noisy linear measurements with unknown parameters.
method Vector approximate message passing (VAMP) combined with Expectation-Maximization (EM).
result EM-VAMP algorithm yields stationary points of a free-energy, providing a variational interpretation.
Sharp-SSL uses random projections to identify important variables for semi-supervised learning.
problem High-dimensional semi-supervised learning problems.
method Careful aggregation of low-dimensional results from many axis-aligned random projections.
result Sharp-SSL algorithm can recover signal coordinates with high probability.
Algorithm learns mixtures of Markov chains and MDPs from short trajectories.
problem Learning mixtures of Markov chains and MDPs from short unlabeled trajectories.
method Subspace estimation, spectral clustering, EM algorithm, model estimation, classification.
result 96.6% average accuracy on a mixture of two MDPs in gridworld, outperforming EM algorithm with random initialization.
The EM algorithm performs well for mixture models of Laplacian distributions.
problem Understanding the behavior of the EM algorithm for mixture models.
method Analysis of a simple mixture of Laplacian distributions and proof of convergence for the EM algorithm.
result The EM algorithm converges to the ground truth parameters almost surely with random initialization.
Random utility theory models an agent's preferences on alternatives by drawing a real-valued score on each alternative (typically independently) from a parameterized distribution, and then ranking the alternatives according to scores. A special case that has received significant attention is the Plackett-Luce model, fo…
Matrix completion, i.e., the exact and provable recovery of a low-rank matrix from a small subset of its elements, is currently only known to be possible if the matrix satisfies a restrictive structural constraint---known as {\em incoherence}---on its row and column spaces. In these cases, the subset of elements is sam…
FIEM accelerates EM for large datasets with nonasymptotic convergence bounds.
problem Efficiently optimizing large datasets using EM framework.
method FIEM recasts EM in Stochastic Approximation framework and provides nonasymptotic convergence bounds.
result Nonasymptotic bounds for convergence in expectation as a function of n and $\kmax$. In this article, we derive a new stepsize adaptation for the normalized least mean square algorithm (NLMS) by describing the task of linear acoustic echo cancellation from a Bayesian network perspective. Similar to the well-known Kalman filter equations, we model the acoustic wave propagation from the loudspeaker to th…
Proposes new methods for Markov chain choice models with panel data.
problem Dependence among transactions for the same customer in historical data.
method Expectation-maximization (EM) algorithms incorporating partial-ordering preference information.
result EM algorithms outperform traditional methods on synthetic and real datasets.