Paper proposes mutual information learning for deep learning classifiers.
problem Overfitting in deep learning models.
method Mutual information learning framework to train classifiers.
result MILCs achieve better generalization than conditional entropy classifiers.
In this paper, we present a new approach to interpret deep learning models. By coupling mutual information with network science, we explore how information flows through feedforward networks. We show that efficiently approximating mutual information allows us to create an information measure that quantifies how much in…
Deep learning model estimates mutual information with low bias and variance.
problem Estimating mutual information between continuous variables.
method Supervised deep learning approach using Linfoot informational correlation as labels.
result Lower bias and variance compared to other methods.
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.
Deep learning models predict mutual funds' performance better than traditional methods.
problem Predicting mutual funds' performance accurately.
method Deep learning models (LSTM, GRUs) trained with Bayesian optimization and ensemble methods.
result Ensemble method of LSTM and GRUs achieves the highest accuracy in forecasting mutual funds' Sharpe ratios.
Proposes MIGE for accurate MI gradient estimation in high-dimensional settings.
problem Intractability of MI in continuous and high-dimensional settings.
method Score estimation of implicit distributions for gradient estimation of MI.
result MIGE provides tight and smooth gradient estimation of MI in high-dimensional settings.
We examine a class of deep learning models with a tractable method to compute information-theoretic quantities. Our contributions are three-fold: (i) We show how entropies and mutual informations can be derived from heuristic statistical physics methods, under the assumption that weight matrices are independent and ort…
This paper demonstrates dynamic hyper-parameter setting, for deep neural network training, using Mutual Information (MI). The specific hyper-parameter studied in this paper is the learning rate. MI between the output layer and true outcomes is used to dynamically set the learning rate of the network through the trainin…
In this work, we perform unsupervised learning of representations by maximizing mutual information between an input and the output of a deep neural network encoder. Importantly, we show that structure matters: incorporating knowledge about locality of the input to the objective can greatly influence a representation's …
Develops a method to explain deep learning models for complex systems.
problem Rapid simulation-based prototyping of complex systems with high-dimensional CVs and QoIs.
method Moment-independent global sensitivity analysis using differential mutual information.
result Surrogate model driven by mutual information provides useful rankings and optimizations.
The paper proposes a method to detect and filter noisy or mislabeled data using pointwise mutual information.
problem Detecting and filtering noisy or mislabeled data in deep learning models.
method A mutual information-based framework quantifying statistical dependencies between inputs and labels.
result The method effectively filters low-quality samples, improving classification accuracy by up to 15%.
Optimal adversarial attacks minimize mutual information, revealing classifier vulnerabilities.
problem Designing optimal attacks to degrade machine learning performance.
method Information-theoretic approach to finding optimal perturbations.
result Optimal attacks minimize mutual information between degraded and original signals.
End-to-end learning of codes for secure BPSK communication in Gaussian wiretap channel.
problem Secure communication in the presence of an adversary for BPSK modulation.
method Learning finite-length codes using deep neural networks (DNNs) with mutual information estimation.
result Learned codes achieve near-optimal security performance, as demonstrated by numerical results.
Symmetry transformations induce invariances which are frequently described with deep latent variable models. In many complex domains, such as the chemical space, invariances can be observed, yet the corresponding symmetry transformation cannot be formulated analytically. We propose to learn the symmetry transformation …
This paper introduces GEMINI, a new mutual information metric for unsupervised neural network training.
problem The mutual information (MI) as a clustering objective does not lead to satisfactory clusters.
method The authors generalised MI by changing its core distance, introducing GEMINIs that do not require regularizations and can automatically select the number of clusters.
result GEMINIs can automatically select the number of clusters without requiring a priori knowledge of the number of clusters.
The mutual information is a core statistical quantity that has applications in all areas of machine learning, whether this is in training of density models over multiple data modalities, in maximising the efficiency of noisy transmission channels, or when learning behaviour policies for exploration by artificial agents…
DHOG improves unsupervised clustering accuracy on image benchmarks.
problem Local optima in mutual information maximisation lead to suboptimal representations.
method Deep hierarchical object grouping (DHOG) computes multiple discrete representations in a hierarchical order.
result DHOG achieves new state-of-the-art results on three main benchmarks.
Discriminative clustering uses mutual information to cluster data.
problem Clustering data into cohesive groups.
method Discriminative clustering using mutual information.
result Mutual information has been a cornerstone of discriminative clustering.
Exploration is a difficult challenge in reinforcement learning and is of prime importance in sparse reward environments. However, many of the state of the art deep reinforcement learning algorithms, that rely on epsilon-greedy, fail on these environments. In such cases, empowerment can serve as an intrinsic reward sign…
Spatio-temporal graphs such as traffic networks or gene regulatory systems present challenges for the existing deep learning methods due to the complexity of structural changes over time. To address these issues, we introduce Spatio-Temporal Deep Graph Infomax (STDGI)---a fully unsupervised node representation learning…
A new algorithm improves efficiency in selecting examples for deep learning.
problem Efficiently choosing multiple examples to mark up for deep learning on large datasets.
method Large BatchBALD algorithm, approximating BatchBALD with reduced computational complexity.
result Comparable quality in selection while significantly reducing computation time, especially for large batches.
In this paper we consider the problem of semi-supervised learning with deep Convolutional Neural Networks (ConvNets). Semi-supervised learning is motivated on the observation that unlabeled data is cheap and can be used to improve the accuracy of classifiers. In this paper we propose an unsupervised regularization term…
Theory proposes neural networks can be initialized for optimal information transmission.
problem Optimizing neural networks for optimal information transmission and representation.
method Developed a corrected mean-field framework to study neural networks as information channels, proving mutual information maximization at dynamic isometry.
result Mutual information maximization is realized between inputs and propagated signals when neural networks are initialized at dynamic isometry.
This work improves independence tests for high-dimensional data.
problem Detecting subtle dependencies between high-dimensional random variables with complex distributions.
method Develops two approaches to learn powerful independence tests using variational mutual information and HSIC.
result Optimized HSIC tests generally outperform other approaches on detecting structured dependence.
This paper improves DNN generalization by accurately estimating mutual information.
problem Intractability of estimating mutual information in DNNs.
method Introduces a probabilistic representation of DNNs to accurately estimate mutual information.
result Derives a tighter generalization bound than previous relaxations.
New method bounds membership inference attack success using mutual information.
problem Vulnerability of deep neural networks to membership inference attacks.
method Extended Fano's inequality to measure mutual information between inputs and activations.
result Empirical evaluation shows strong correlation between mutual information and model susceptibility.
New method learns diverse solutions in reinforcement learning without gradient bias.
problem Lack of diverse solutions in reinforcement learning tasks.
method Maximizes state-action-based mutual information directly, using variational lower bound.
result Successfully learns an infinite set of diverse solutions.
We present a representation learning framework for financial time series forecasting. One challenge of using deep learning models for finance forecasting is the shortage of available training data when using small datasets. Direct trend classification using deep neural networks trained on small datasets is susceptible …
Neural estimator improves mutual information estimation in high dimensions.
problem Estimating mutual information in high dimensions is challenging.
method Parametrizing conditional densities with normalizing flows and using block autoregressive structure.
result Improved mutual information estimation on benchmark tasks.
New framework improves multivariate time series forecasting by minimizing redundant information.
problem Improving multivariate time series forecasting with deep learning techniques.
method Cross-variable Decorrelation Aware feature Modeling (CDAM) and Temporal correlation Aware Modeling (TAM) to refine Channel-mixing and exploit temporal correlations.
result Significantly surpasses existing models in comprehensive tests.
Traditional anatomical analyses captured only a fraction of real phenomic information. Here, we apply deep learning to quantify total phenotypic similarity across 2468 butterfly photographs, covering 38 subspecies from the polymorphic mimicry complex of Heliconius erato and Heliconius melpomene. E…
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.
Novel bounds for deep MDA algorithms improve performance and efficiency.
problem Improving performance of MDA algorithms with few target labels and pseudo labels.
method Information-theoretic tools and novel deep MDA algorithm.
result Algorithm-dependent generalization bounds for MDA.
A method to improve image synthesis diversity using mutual information.
problem Mode collapse in conditional GANs for multimodal image synthesis.
method Explicitly estimate and maximize mutual information between latent code and output image.
result Prevents mode collapse and encourages synthesis of diverse images.
Cross-entropy loss linked to metric learning, outperforming complex pairwise losses.
problem Improving metric learning performance without complex optimization schemes.
method Theoretical analysis linking cross-entropy to pairwise losses, showing cross-entropy as an upper bound and equivalent to mutual information maximization.
result Minimizing cross-entropy is equivalent to maximizing mutual information, leading to state-of-the-art performance.
Bounding the generalization error of learning algorithms has a long history, which yet falls short in explaining various generalization successes including those of deep learning. Two important difficulties are (i) exploiting the dependencies between the hypotheses, (ii) exploiting the dependence between the algorithm'…
We describe a novel extension of soft actor-critics for hierarchical Deep Q-Networks (HDQN) architectures using mutual information metric. The proposed extension provides a suitable framework for encouraging explorations in such hierarchical networks. A natural utilization of this framework is an adversarial setting, w…
New bounds improve neural network generalization through slicing.
problem Difficulty in evaluating mutual information in high dimensions for neural networks.
method Slicing the parameter space and using disintegrated mutual information and k-sliced mutual information.
result Slicing improves generalization and offers significant computational and statistical advantages.
We propose a new framework for reasoning about information in complex systems. Our foundation is based on a variational extension of Shannon's information theory that takes into account the modeling power and computational constraints of the observer. The resulting \emph{predictive V-information} encompasse…
Deep learning models can be understood through information theoretic analysis.
problem Understanding the learned representations of neural networks in security contexts.
method Information theoretic analysis of neural network representations using mutual information and homeomorphism.
result Mutual information remains invariant under homeomorphism, suggesting that neural networks require meaningful feature engineering.
This paper introduces GEMINI, a new metric for unsupervised neural network training that avoids the need for regularizations.
problem The mutual information (MI) as a clustering objective does not lead to satisfactory clusters.
method The authors generalised the mutual information by changing its core distance, introducing the Generalised Mutual Information (GEMINI).
result Some GEMINIs do not require regularizations when training and can automatically select the number of clusters.
A novel algorithm for unsupervised graph representation learning combining coarsening and mutual information maximization.
problem Current limitations in unsupervised graph representation learning, especially in embedding new graphs and considering both micro- and macro-structures.
method Combines coarsening with mutual information maximization to produce high-quality embeddings.
result The algorithm produces high-quality embeddings that are competitive with state-of-the-art methods.
Deep nonlinear models pose a challenge for fitting parameters due to lack of knowledge of the hidden layer and the potentially non-affine relation of the initial and observed layers. In the present work we investigate the use of information theoretic measures such as mutual information and Kullback-Leibler (KL) diverge…
ResNet learns to compress information during training.
problem Applying information bottleneck principle to deep learning models.
method Trained PixelCNN++ models to measure mutual information between ResNet hidden layers and input data.
result Compression occurs in both classification and autoencoding training regimes.
We introduce Minimal Achievable Sufficient Statistic (MASS) Learning, a training method for machine learning models that attempts to produce minimal sufficient statistics with respect to a class of functions (e.g. deep networks) being optimized over. In deriving MASS Learning, we also introduce Conserved Differential I…
NCFA uses deep learning and causal discovery to analyze complex data.
problem Analyzing complex, interdependent data with causal relationships.
method NCFA combines latent causal discovery and variational autoencoders.
result NCFA outperforms standard VAEs in sparsity, complexity, and causal interpretability.
In the context of recent deep clustering studies, discriminative models dominate the literature and report the most competitive performances. These models learn a deep discriminative neural network classifier in which the labels are latent. Typically, they use multinomial logistic regression posteriors and parameter re…
SepIt improves speech separation for multiple speakers.
problem Improving speech separation for multiple speakers in single channel recordings.
method SepIt uses a deep neural network that iteratively improves estimates of different speakers based on mutual information.
result SepIt outperforms state-of-the-art methods for 2, 3, 5, and 10 speakers.