Three LF training criteria improve neural network acoustic models without cross-entropy pre-training.
problem Improving purely sequence-trained neural network acoustic models.
method Comparison of three lattice-free discriminative training criteria (MMI, bMMI, sMBR) on LVCSR tasks.
result LF-bMMI models outperform plain LF-MMI models by 5% WER on Switchboard datasets.
We find the maximum mutual information for neural networks and its key determinants.
problem Understanding the maximum mutual information in neural architectures.
method Derived closed-form expression for maximum mutual information across neural network families.
result Maximum mutual information stems from a generalized formula and is influenced by network width and statistical invariances.
A new regularizer boosts long-range dependency in sequence data.
problem Improving long-range dependency in sequence data models.
method Developed a mutual information regularizer to enhance sequence learning.
result The approach increases mutual information and likelihood on holdout data.
SURI boosts features with high unique relevant information for better health data analysis.
problem Preserving interpretability in health data analysis.
method Mutual information-based feature selection (MIBFS) method called SURI.
result SURI selects more relevant features leading to higher classification performance.
Williams and Beer (2010) proposed a nonnegative mutual information decomposition, based on the construction of redundancy lattices, which allows separating the information that a set of variables contains about a target variable into nonnegative components interpretable as the unique information of some variables not p…
Exponential models of distributions are widely used in machine learning for classiffication and modelling. It is well known that they can be interpreted as maximum entropy models under empirical expectation constraints. In this work, we argue that for classiffication tasks, mutual information is a more suitable informa…
MIM learns useful representations with high mutual information.
problem Learning useful representations for downstream tasks.
method Symmetric Jensen-Shannon divergence and mutual information regularizer in an encoder/decoder framework.
result MIM learns high mutual information representations without posterior collapse.
MIM learns joint distributions with mutual information and low divergence.
problem Learning joint distributions over observations and latent variables.
method Probabilistic auto-encoder with three design principles: low divergence, high mutual information, and low marginal entropy.
result MIM learns representations with high mutual information, consistent encoding and decoding distributions, effective latent clustering, and comparable data log likelihood to VAE.
Novel mutual information bound improves statistical inference rates.
problem Improving statistical inference rates in Bayesian nonparametrics.
method Introduces a novel mutual information bound.
result Improved contraction rates for fractional posteriors.
CM algorithm improves MMI classifications for unseen instances.
problem Improving classification accuracy for unseen instances using MMI criterion.
method Introduces CM algorithm for MMI classifications, combining semantic and Shannon channels for matching.
result Achieves high mutual information (99%) with minimal iterations in low-dimensional feature spaces.
The paper tackles adversarial robustness by maximizing worst-case mutual information.
problem Training robust machine learning models against adversarial inputs is challenging.
method Develops a notion of representation vulnerability and an unsupervised learning method to maximize worst-case mutual information.
result Proves a lower bound on minimum adversarial risk and supports robustness of representations.
InfoMax-VAE enhances VAEs for better representation learning.
problem Lack of meaningful learned representations in VAEs.
method Combines VAEs with mutual information maximization.
result Significantly boosts quality of learned high-level representations.
Paper formulates mutual information optimal control for discrete-time systems.
problem Optimal control of discrete-time linear systems with mutual information.
method Formulates MIOCP as an extension of MEOCP, derives optimal policy and prior, proposes alternating minimization algorithm.
result Proposes an alternating minimization algorithm for MIOCP.
MIND estimates mutual information from ordinal data without full distributional knowledge.
problem Estimating mutual information from ordinal data with limited data.
method Copula-based maximum-entropy estimation of copula entropies.
result MIND estimator is consistent, data-efficient, and unbounded for any sample size.
A new method for automatic gradient tree boosting using information theory.
problem Automatic selection of tree complexity and number in gradient boosting.
method Optimism of greedy leaf splitting procedure modeled as a Cox-Ingersoll-Ross process, leading to an information criterion for model selection.
result The method achieves significant speedups (10-1400) compared to xgboost without sacrificing predictive power.
Mean field Gaussian inference limits mutual information to regularize neural networks.
problem Understanding and quantifying the regularization effect of mean field Gaussian inference.
method Empirically observed and theoretically quantified mutual information limitation through noise.
result Bounding mutual information between parameters and data effectively regularizes neural networks.
A neural network approach for feature selection using mutual information.
problem Feature ranking and selection leading to sub-optimal solutions for class separability.
method Stochastic mutual information gradient estimation for dimensionality reduction.
result The network projects features onto an output space maximizing mutual information with class labels.
Improves data efficiency for estimating mutual information.
problem Estimating mutual information from limited data.
method Developed a Data-Efficient MINE Estimator (DEMINE) and Meta-DEMINE.
result Significantly improved data efficiency in estimating mutual information.
AMI framework improves text generation by optimizing mutual information between source and target.
problem Previous MI approaches ignored the backward network, leading to loose variational bounds.
method AMI is a saddle point optimization framework that iteratively promotes and demotes generated instances.
result AMI significantly outperforms baselines on various text generation tasks.
cMIM improves representation learning without positive-pair augmentations.
problem Learning robust representations for diverse tasks.
method Contrastive Mutual Information Machine (cMIM) framework.
result cMIM outperforms MIM and InfoNCE on classification and regression tasks.
MSRL learns a representation maximizing mutual info with response variables.
problem Learning sufficient representations for complex, multi-dimensional data.
method Variational mutual information, deep neural networks, generalized Dudley's inequality.
result MSRL achieves consistent and accurate representation learning.
Novel approach uses neural networks to enhance CI testing for feature selection.
problem Challenges in implementing Markov blanket feature selection due to CI testing limitations.
method Two-step approach: feature mapping followed by CI testing using k-NN. result The method boosts CI testing performance, leading to improved feature selection.
Neural networks help create summary statistics for complex models.
problem Creating summary statistics for models with intractable likelihood functions.
method Infomax learning with neural networks to maximize mutual information.
result Improves performance of approximate Bayesian computation and neural likelihood methods.
We discuss the connection between information and copula theories by showing that a copula can be employed to decompose the information content of a multivariate distribution into marginal and dependence components, with the latter quantified by the mutual information. We define the information excess as a measure of d…
New bounds for optimal transport using Gaussian processes and rate-distortion functions.
problem Finding bounds for entropic optimal transport with mutual information constraints.
method Lifting technique to construct a Gaussian process and applying the majorizing measure theorem.
result Maximum expected inner product is equivalent to a truncated integral involving the rate-distortion function.
In this paper we analyze boosting algorithms in linear regression from a new perspective: that of modern first-order methods in convex optimization. We show that classic boosting algorithms in linear regression, namely the incremental forward stagewise algorithm (FSε) and least squares boosting (LS-Boost($…
Develops LSH schemes for f-divergences and mutual information loss.
problem Approximating nearest neighbors in high-dimensional probability distributions.
method General framework and specific LSH schemes for f-divergences and mutual information loss.
result Generalized Jensen-Shannon divergence can be approximated by Hellinger distance.
A boosting method improves nonparametric density estimation without smoothing assumptions.
problem Overfitting in nonparametric data fitting.
method Introduces a boosting algorithm for univariate nonparametric maximum likelihood estimation.
result Demonstrates the effectiveness of the boosting approach through simulations and real data experiments.
The Partial Information Decomposition (PID) [arXiv:1004.2515] provides a theoretical framework to characterize and quantify the structure of multivariate information sharing. A new method (Idep) has recently been proposed for computing a two-predictor PID over discrete spaces. [arXiv:1709.06653] A lattice of maximum en…
We define On-Average KL-Privacy and present its properties and connections to differential privacy, generalization and information-theoretic quantities including max-information and mutual information. The new definition significantly weakens differential privacy, while preserving its minimalistic design features such …
This article describes the R package varrank. It has a flexible implementation of heuristic approaches which perform variable ranking based on mutual information. The package is particularly suitable for exploring multivariate datasets requiring a holistic analysis. The core functionality is a general implementation of…
In data science, it is often required to estimate dependencies between different data sources. These dependencies are typically calculated using Pearson's correlation, distance correlation, and/or mutual information. However, none of these measures satisfy all the Granger's axioms for an "ideal measure". One such ideal…
Forecastability measures predictive information across horizons.
problem How much predictive information is available at each prediction horizon?
method Develops the consequences of mutual information between future observations and information set.
result Forecastability is a profile reflecting process dependence structure, with properties like compression and truncation error.
Meta Fusion integrates various multimodal data fusion strategies into a unified framework.
problem Improving predictive power of machine learning methods across diverse applications.
method Meta Fusion constructs a cohort of models based on latent representations across modalities, sharing soft information to boost performance.
result Meta Fusion consistently outperforms conventional fusion strategies in simulation and real-world applications.
DeGNN improves graph neural networks by decomposing graphs.
problem Graph Convolutional Networks (GCNs) suffer from oversmoothing and limited depth.
method Characterized oversmoothing through information theory, proposed DeGNN for automatic graph decomposition.
result DeGNN boosts performance of general GNNs and achieves state-of-the-art results.
Boosting improves ICA for better component recovery.
problem Improving ICA's reliance on prior knowledge of sources.
method Maximizing likelihood via boosting and fixed-point unmixing.
result Boosting-based ICA outperforms existing methods.
Study entropic regularization of Gaussian measures and processes on Hilbert space.
problem Regularizing 2-Wasserstein distance for infinite-dimensional Gaussian measures and processes.
method Minimum Mutual Information property, closed form formulas, Fréchet differentiability, Sinkhorn barycenter equation.
result Entropic 2-Wasserstein distance and Sinkhorn divergence are Fréchet differentiable in Hilbert space.
This paper resolves the all-or-nothing phase transition in graph matching.
problem Recovering vertex correspondence between edge-correlated random graphs.
method Analysis of mutual information, truncated second-moment computation, and maximum likelihood estimator.
result Sharp thresholds for correct matching in both dense and sparse graphs.
New method tightens bounds on causation probabilities using independent datasets.
problem Challenging point identification of causation probabilities without strong assumptions.
method Imposes counterfactual consistency between SCMs constructed from independent datasets and uses conditional mutual information.
result Significantly tighter bounds on causation probabilities are established.
SMOTE is one of the oversampling techniques for balancing the datasets and it is considered as a pre-processing step in learning algorithms. In this paper, four new enhanced SMOTE are proposed that include an improved version of KNN in which the attribute weights are defined by mutual information firstly and then they …
Proposes new feature transformation methods for brain interface models.
problem Sub-optimality of feature ranking and selection in brain interface models.
method Introduces maximum mutual information linear and nonlinear transformations.
result Significantly better performance in binary and multi-class decoding analyses.
New method calculates empowerment from visual data, solving complex reinforcement learning problems.
problem Calculating empowerment in unknown dynamics from visual observation is challenging.
method Developed a novel approach using stochastic dynamic models in latent space and the Water-Filling algorithm.
result Efficiently computed empowerment in unknown dynamics from visual observation only.
The paper proves mutual information measurement is statistically limited.
problem Measuring mutual information from finite data is difficult.
method Proves statistical limitations on any method of measuring mutual information.
result Any distribution-free high-confidence lower bound on mutual information estimated from N samples cannot be larger than O(ln N ).
The paper tackles model selection for unseen tasks by capturing relationships among checkpoints.
problem Deciding which model combinations are likely to be effective for a new task is difficult.
method The paper models the task space as a Gaussian process and identifies representative checkpoints using mutual information and a greedy algorithm.
result Representative checkpoints generalize to new tasks with superior performance.
Improved binning technique boosts nUV measure performance.
problem Improving the performance of the nUV measure in real applications.
method Introduced the nUV measure, provided theoretical optimal binning techniques, and proposed algorithms for approximate solutions.
result Approximate binning techniques show 4-13% increase in AUC scores with statistical significance.
New neural network approach using mutual information.
problem Training neural networks for imbalanced datasets.
method Converts neural network classifiers to mutual information evaluators.
result New form of softmax leads to better classification accuracy, especially for imbalanced datasets.
We consider the problem of universal joint clustering and registration of images and define algorithms using multivariate information functionals. We first study registering two images using maximum mutual information and prove its asymptotic optimality. We then show the shortcomings of pairwise registration in multi-i…
Maximum likelihood is the most widely used statistical estimation technique. Recent work by the authors introduced a general methodology for the construction of estimators for functionals in parametric models, and demonstrated improvements - both in theory and in practice - over the maximum likelihood estimator (MLE), …