Bayesian sparsification reduces memory usage in RNNs without significant loss of accuracy.
problem High memory usage in RNNs for text analysis tasks.
method Apply Sparse Variational Dropout and Binary Variational Dropout to sparsify RNNs.
result Achieve 99.5% sparsity on sentiment analysis with no quality drop and up to 87% on language modeling with slight loss of accuracy.
Variational Dropout sparsifies deep neural networks by reducing parameter count.
problem Overfitting and high parameter count in deep neural networks.
method Extending Variational Dropout to unbounded dropout rates, reducing gradient variance, and applying individual dropout rates per weight.
result Significant reduction in parameter count (up to 280 times) with minimal accuracy loss.
QSD enhances deep network performance through biologically plausible dropout.
problem Overfitting in deep networks.
method Quantal Synaptic Dilution (QSD) model based on neuronal synapses.
result QSD outperforms standard dropout in various deep network architectures.
Adaptive dropout and regularization are shown to be dual in linear networks.
problem Sparsifying deep neural networks.
method Examining dropout in the linear case, revealing a duality with regularization.
result Adaptive dropout methods lead to sparse solutions with effective penalties similar to classical sparse optimization penalties.
This research explores using Alpha-Divergences in variational dropout for better inference.
problem Improving variational inference methods using alternative divergences.
method Extending the Stochastic Gradient Variational Bayes (SGVB) framework with Alpha-Divergences.
result The α-divergence with αightarrow1 yields the lowest training error and optimizes the ELBO. Variational dropout and ARD are shown to be equivalent, improving sparsity in neural networks.
problem Improper priors in variational dropout models and theoretical issues.
method Applying ARD to Bayesian DNNs with Gaussian approximations and exploring hierarchical priors.
result Variational dropout and ARD objectives are equivalent, with improved sparsity.
Privacy-preserving dropout technique for neural networks.
problem Overfitting in deep neural networks and privacy of training data.
method Modified variational dropout with concentrated differential privacy.
result The technique provides differential privacy for neural networks.
Improves Bayesian neural networks inference efficiency and accuracy.
problem Inflexibility of factorized structure in Dropout posterior.
method Introduces Variational Structured Dropout (VSD) with orthogonal transformation.
result VSD induces adaptive regularization and better generalization.
The paper calibrates uncertainty in dropout variational inference models.
problem Miscalibration of model uncertainty in dropout variational inference.
method Logit scaling methods are extended to recalibrate model uncertainty.
result Logit scaling reduces miscalibration, improving reliability of predictions.
Adaptive network sparsification improves model compactness and accuracy.
problem Suboptimal network sparsification due to input-independent dropout.
method Dependent variational beta-Bernoulli dropout.
result Significantly more compact networks with consistent accuracy improvements.
This paper calibrates uncertainty in dropout variational inference models.
problem Uncertainty in variational inference with dropout is poorly calibrated.
method Temperature scaling is extended to dropout variational inference.
result Temperature scaling reduces miscalibration of uncertainty.
We investigate a local reparameterizaton technique for greatly reducing the variance of stochastic gradients for variational Bayesian inference (SGVB) of a posterior over model parameters, while retaining parallelizability. This local reparameterization translates uncertainty about global parameters into local noise th…
VBD improves variational dropout by using a hierarchical prior, enabling better regularization.
problem Improper log-uniform prior in VD causes ill-posed posterior inference.
method Introduces a hierarchical prior with a zero-mean Gaussian distribution and a uniform hyper-prior.
result VBD enables well-posed posterior inference and superior regularization performance.
Stochastic Sparse Subspace Clustering improves subspace clustering by reducing over-segmentation through dropout.
problem Over-segmentation in subspace clustering.
method Introducing dropout regularization to enforce denser connections between points from the same subspace.
result Stochastic Sparse Subspace Clustering effectively handles large datasets and reduces over-segmentation.
Researchers fix issues with variational Bayesian dropout.
problem Variational Bayesian dropout's theoretical framework has issues.
method Proposed new approximate inference objective called Quasi-KL (QKL).
result QKL addresses singularity issue and leads to Principal Component Analysis solution.
Advanced Dropout improves DNN performance without requiring model-specific dropout techniques.
problem Overfitting in deep neural networks due to lack of data.
method Model-free, end-to-end training with adaptive dropout rate optimization.
result Advanced Dropout outperforms other techniques on various datasets.
NADPEx uses dropout to enable temporally consistent exploration in reinforcement learning.
problem Achieving temporally consistent exploration in reinforcement learning agents.
method Integrates dropout into reinforcement learning policies to ensure temporal consistency.
result NADPEx outperforms naive exploration and parameter noise in tasks with sparse rewards.
Proposes LBD for more robust deep learning models.
problem Improving robustness and uncertainty in deep learning models.
method Model-agnostic learnable Bernoulli dropout with joint optimization of dropout rates.
result Superior performance compared to other dropout schemes.
Bayesian inference using stochastic neural networks ensembles.
problem Approximating Bayesian posterior distributions.
method Formulate stochastic ensembles of neural networks, train with variational inference, and evaluate using Monte Carlo dropout.
result Stochastic ensembles provide more accurate posterior estimates than other methods.
DropLasso improves scRNA-seq data analysis by handling dropout noise.
problem Noisy scRNA-seq data with many dropout events.
method Extends dropout regularization to sparse linear models.
result DropLasso provides better molecular signature inference than standard regularizations.
MC Dropout is re-evaluated as not Bayesian, affecting predictive posterior and multimodality.
problem MC Dropout's Bayesian properties and predictive posterior are questioned.
method Re-evaluation of MC Dropout's properties, including a new VI engine in pytorch.
result MC Dropout does not produce a faithful Bayesian predictive posterior.
New Bayesian model injects noise to improve neural network sparsity and acceleration.
problem Improving neural network sparsity and acceleration.
method Proposes a new Bayesian model that injects noise to neurons outputs while keeping weights unregularized, using log-normal multiplicative noise.
result Provides significant acceleration on deep neural architectures.
Recurrent neural networks (RNNs) stand at the forefront of many recent developments in deep learning. Yet a major difficulty with these models is their tendency to overfit, with dropout shown to fail when applied to recurrent layers. Recent results at the intersection of Bayesian modelling and deep learning offer a Bay…
Analyzes dropout in neural networks using differential equations.
problem Lack of theoretical understanding of dropout's effectiveness.
method Analytical study of dropout in two-layer neural networks with online stochastic gradient descent.
result Dropout reduces detrimental correlations and optimal probability increases with data noise.
New method reduces parameter overhead for Bayesian neural networks.
problem High parameter overhead and difficulty of implementation in variational Bayesian neural networks.
method Constructs a general variational family for ensemble-based Bayesian neural networks that works well with batch normalization layers.
result Improves predictive accuracy and achieves almost perfect calibration on a ResNet-18 trained with ImageNet.
A method to train neural networks that are robust to pruning.
problem Training neural networks to be amenable to pruning.
method Targeted dropout: a simple self-reinforcing sparsity criterion to select units or weights to be dropped.
result Trained networks are robust to post hoc pruning of weights or units.
Improved dropout inference for Bayesian neural networks using alpha-divergences.
problem Uncertainty underestimation in dropout variational inference.
method Proposed a re-parametrisation of alpha-divergence objectives for dropout networks.
result Improved uncertainty estimates and accuracy compared to VI in dropout networks.
Batch normalisation doesn't affect variational inference but fails for larger batch sizes.
problem Failure of Monte Carlo Batch Normalisation (MCBN) for capturing epistemic uncertainty in larger batch sizes.
method Investigated MCBN as an approximate inference technique for Bayesian neural networks, showing its limitations and providing insights for improvement.
result For larger batch sizes, MCBN fails to capture epistemic uncertainty, requiring the batch size to be a variational parameter.
A new loss function using noisy activations improves deep learning performance.
problem Improving deep learning performance through optimal representation learning.
method Regularization with multiplicative noise in activations, Information Dropout.
result Information Dropout achieves comparable or better generalization performance than binary dropout.
A new neural network architecture reduces parameters by 94% while maintaining performance.
problem Reduction of trainable parameters in neural networks.
method Spatially-coupled sparse construction to allocate trainable parameters efficiently.
result Performance comparable to traditional neural networks with 94% fewer parameters.
Dropout has recently emerged as a powerful and simple method for training neural networks preventing co-adaptation by stochastically omitting neurons. Dropout is currently not grounded in explicit modelling assumptions which so far has precluded its adoption in Bayesian modelling. Using Bayesian entropic reasoning we s…
VINNAS uses variational inference to avoid mode collapse in neural architecture search.
problem Mode collapse in gradient-based NAS methods, leading to suboptimal architectures.
method Differentiable variational inference with variational dropout and automatic relevance determination.
result State-of-the-art accuracy with up to twice fewer non-zero parameters.
Enhances uncertainty estimation in medical image segmentation.
problem Frequency-related noise in medical imaging leads to biased uncertainty estimates.
method Extends MC-Dropout to the frequency domain for better uncertainty estimation.
result MC-Frequency Dropout improves calibration and uncertainty in semantic segmentation.
Evaluates uncertainty quality in neural networks using anomaly detection.
problem Evaluating the quality of uncertainty in neural networks.
method Extract uncertainty measures, use them as features for an anomaly detector, and compare different neural network models.
result Bayesian Dropout and OSBA provide better uncertainty information than Maximum Likelihood, and are faster.
Proposes using Monte Carlo Dropout in Autoencoder and VAE for synthetic data generation.
problem Handling large amounts of data in costly or difficult-to-collect scenarios.
method Incorporates Monte Carlo Dropout within Autoencoder and Variational Autoencoder.
result Generated data sets are statistically and predictively similar to actual data.
Prb-GAN uses dropout and variational inference to improve GAN performance.
problem GANs struggle with mode loss and training instability.
method Introduces Prb-GANs with dropout and variational inference for parameter distribution.
result Improves GAN performance through dropout and variational inference.
Dropout improves matrix factorization by controlling factor size.
problem Understanding regularization properties of dropout for matrix factorization.
method Theoretical analysis of dropout's equivalence to a deterministic model with adaptive dropout rates.
result Dropout's regularization effect is limited by the fixed dropout rate, suggesting adaptive rates.
Novel framework improves deep learning uncertainty estimates.
problem Improving uncertainty estimates in deep learning models.
method Statistical mechanics based framework for dropout, focusing on variance of loss.
result Generates don't-know answers in classification tasks, improving classifier reliability.
New method improves neural network compression.
problem Efficiently compressing neural networks for mobile devices and inference.
method Combining Soft-Weight Sharing and Variational Dropout.
result New approach achieves state-of-the-art results in model compression.
Model predicts higher education dropout risk with interpretable parameters.
problem Predicting and understanding student dropout risk in higher education.
method Sparse interpretable post-clustering logistic regression.
result Model identifies distinct dropout risk subgroups within the student population.
EB-VAE combines tumor growth and dropout data for personalized treatment response modeling.
problem Challenges in integrating longitudinal tumor measurements, dropout information, and genetic covariates.
method Extended EB-VAE framework to jointly model longitudinal and time-to-event data, incorporating dropout hazard and genetic covariates.
result Hybrid decoder formulation yields consistent treatment-effect parameters and prior predictive performance comparable to neural decoder.
Bayesian approach improves activity recognition accuracy and uncertainty quantification.
problem Lack of predictive uncertainty in multimodal audiovisual activity recognition.
method Uncertainty aware multimodal Bayesian fusion framework combining deterministic and variational layers.
result Improved precision-recall AUC by 10.2% on MiT dataset.
Bayesian sparsification improves complex-valued neural networks by 50-100x with minimal performance loss.
problem Efficiently compressing complex-valued neural networks for embedded systems.
method Extending Sparse Variational Dropout to complex-valued networks and conducting a numerical study.
result Achieved state-of-the-art performance on MusicNet with 50-100x compression.
Paper proposes a new method for imputing missing biomedical data.
problem Missing values in biomedical data.
method Monte Carlo Dropout within Autoencoders.
result The method improves imputation error and predictive similarity.
A fast and scalable method for variable selection in high-dimensional Gaussian processes.
problem Inefficient variable selection in high-dimensional Gaussian processes.
method Developed a fast and scalable variational inference algorithm for spike and slab Gaussian processes.
result Consistently outperforms vanilla and sparse variational GPs while retaining similar runtimes.
Novel regularization for Vision Transformers improves model generalization and sparsity.
problem Improving generalization and sparsity in Vision Transformers.
method Likelihood-guided variational Ising-based regularization.
result Improved generalization and sparsity in Vision Transformers.
Detects out-of-distribution sentences in Neural Machine Translation.
problem Identifying sentences from a different language than the training data.
method Developed a new uncertainty measure for long sequences of words in Transformers.
result Shows ability to identify Dutch sentences as German input.
Radial BNNs offer a scalable, continuous weight distribution for Bayesian deep learning.
problem Discrete support in Bayesian deep learning methods like MC dropout.
method Radial BNNs with full support over weight-space.
result Radial BNNs outperform discrete-support methods in real-world applications.