Review of priors in Bayesian deep learning models.
problem The importance of prior choices in Bayesian deep learning models.
method Overview of different priors and methods of learning priors from data.
result Motivate practitioners to think carefully about prior specification.
Reduces GAN image priors' representation error using a Deep Decoder.
problem Representation error in GAN priors for in-distribution and out-of-distribution images.
method Hybrid model combining GAN prior and Deep Decoder.
result Consistently higher PSNRs on in-distribution and out-of-distribution images.
Deep audio prior uses neural networks to solve audio problems without data.
problem Challenging audio problems like source separation, editing, and synthesis.
method Randomly-initialized neural network with carefully designed audio prior.
result Superior audio results on Universal-150 benchmark dataset.
This paper examines how the choice of prior distribution affects likelihoods of out-of-distribution inputs in deep generative models.
problem Mismatch between prior and data distributions causes deep generative models to assign higher likelihoods to out-of-distribution inputs.
method Proposes using a mixture distribution as a prior to make likelihoods of out-of-distribution inputs more sensitive.
result A mixture prior lowers the out-of-distribution likelihood with respect to real image data sets.
Proposes MOPED method for choosing priors in Bayesian DNNs.
problem Challenges in specifying meaningful priors for deep neural networks.
method Two-stage hierarchical modeling with empirical Bayes.
result MOPED enables scalable variational inference and reliable uncertainty quantification.
Paper improves speech separation by using deep neural networks for more accurate density priors.
problem Improving the accuracy of source priors for independent vector analysis in speech separation.
method Estimating the derivative of speech density using deep neural networks to optimize performance indices.
result Neural network density priors outperform previous ones in convergence speed and SIR.
We introduce a model-based reconstruction framework with deep learned (DL) and smoothness regularization on manifolds (STORM) priors to recover free breathing and ungated (FBU) cardiac MRI from highly undersampled measurements. The DL priors enable us to exploit the local correlations, while the STORM prior enables us …
This paper uses reference priors to improve deep learning models with unlabeled and labeled data.
problem Improving deep learning models with limited labeled data and unlabeled data from the same or related tasks.
method Develops and applies generalizations of reference priors for deep networks to exploit unlabeled and labeled data.
result Demonstrates new semi-supervised learning and pretraining methods for transfer learning.
Bayesian approach improves deep image prior for image reconstruction.
problem Improving performance of deep image prior for image reconstruction tasks.
method Derive Bayesian approach using stochastic gradient Langevin, showing asymptotic equivalence to Gaussian process prior.
result Improves denoising and impainting results for image reconstruction tasks.
This work tackles the challenge of Bayesian deep learning by proposing a new framework for matching Gaussian process priors with neural network parameters.
problem The challenge of specifying priors over neural network parameters, which affects the induced functional prior and is uncontrolled.
method The approach involves defining functional priors using Gaussian processes and matching these priors with the functional prior of neural networks through the minimization of Wasserstein distance.
result The proposed framework offers systematic performance improvements over alternative priors and approximate Bayesian deep learning approaches.
DS2CF-Net learns hierarchical representations with deep coupled factorization and enriched prior.
problem Learning deep hierarchical representations from data.
method Dual-constrained Deep Semi-Supervised Coupled Factorization Network (DS2CF-Net) with enriched prior.
result DS2CF-Net achieves state-of-the-art performance in representation learning and clustering.
Stabilizes deep Bayesian neural networks with self-stabilizing priors.
problem Brittleness and difficulty in training deep Bayesian neural networks.
method Signal propagation theory, reformulated ELBO, self-stabilizing priors.
result Improved convergence and robustness in training deeper networks and noisier settings.
Deep Gaussian Processes with polynomial kernels can collapse rapidly without proper hyperparameter tuning.
problem The collapse of Deep Gaussian Processes with polynomial kernels without careful hyperparameter tuning.
method Analysis using the Berry-Esseen Theorem and observation of prior behavior.
result The prior of a Deep Gaussian Process collapses rapidly towards zero or places negligible mass on low norm functions without proper hyperparameter tuning.
Deep convolutional networks have become a popular tool for image generation and restoration. Generally, their excellent performance is imputed to their ability to learn realistic image priors from a large number of example images. In this paper, we show that, on the contrary, the structure of a generator network is suf…
Bayesian convolutional deep sets improve ambiguity in stationary process modeling.
problem Ambiguity in translation equivariant functional representations due to insufficient data points.
method Introduce Bayesian convolutional deep sets with task-dependent stationary prior.
result Improves representation quality compared to kernel smoother and non-parametric models.
Bayesian method uses deep learning prior for CT reconstruction.
problem Imaging inverse problems in CT reconstruction.
method SA-Roundtrip prior with HMC-pCN sampler.
result Outperforms state-of-the-art methods in CT reconstruction.
Bayesian deep learning uses function-space priors to improve model uncertainty and robustness.
problem Bayesian deep learning struggles with model-specific weight-space priors that are hard to interpret and specify.
method Apply a Dirichlet prior in predictive space and perform approximate function-space variational inference.
result The approach improves uncertainty quantification, scalability, and adversarial robustness in large-scale image classification.
Framework evaluates the impact of prior knowledge in deep learning models.
problem Mitigating data-driven model shortcomings like data dependence and generalization ability.
method Model-agnostic framework inspired by interpretable machine learning, assessing data volume and estimation range effects.
result Complex relationship between data and knowledge, including dependence, synergistic, and substitution effects.
SIGMA prior enables federated learning for non-factorizable models.
problem Current FL methods assume conditional independence, limiting applicability to non-factorizable models.
method SIGMA prior approximates deep generative model to induce conditional independence structure.
result SIGMA prior expands FL applicability to fields requiring modeling dependencies.
Improved computed tomography reconstruction with deep learning and deep image prior.
problem Low data efficiency in computed tomography reconstruction.
method Combining learned primal-dual methods with deep image prior for improved quality and generalization.
result Proposed methods outperform state-of-the-art in low data regime.
Improved deep learning models using new attribution priors and expected gradients.
problem Improving interpretability and performance of deep learning models.
method Introducing new attribution priors and expected gradients method that satisfies interpretability axioms.
result Improves model performance across various real-world tasks.
Survey of deep learning methods for inverse problems, highlighting theoretical challenges.
problem Addressing theoretical gaps in deep learning for inverse problems.
method Survey of existing theoretical developments and open problems.
result Highlighting ongoing challenges in deep learning for inverse problems.
Proposes a method to improve weakly supervised image segmentation using deep geodesic priors.
problem Limited availability of high-quality annotations for image segmentation, especially in medical data.
method Integrates a deep geodesic prior extracted from an auto-encoder to reduce the adverse effects of weak labels in segmentation accuracy.
result The proposed method significantly improves segmentation accuracy, boosting performance by 4.4% in dice score for clean labels and up to 6.3% for noisy labels (L2).
Deep learning and prior maps improve traffic light recognition for autonomous cars.
problem Recognizing traffic lights for autonomous cars in urban environments.
method Combining deep learning-based detection with prior maps for traffic light identification and state recognition.
result The proposed system correctly identified relevant traffic lights along predefined routes.
Method recovers complex-valued signals from speckle-noised measurements.
problem Recovering complex-valued signals from speckle-noised measurements.
method Bagged Deep Image Priors integrated with projected gradient descent and Newton-Schulz algorithm.
result Achieves state-of-the-art performance in MSE reduction.
Improved MRI head anatomy segmentation using deep learning with multiple priors.
problem Challenges in segmenting head anatomy in MRI, especially with lesions.
method Added three types of prior information to a 3D convolutional network: spatial priors, morphological priors, and spatial context.
result Multiprior network improves segmentation performance, especially for abnormal anatomies.
Deep ensembles mimic Bayesian averaging with learned priors.
problem Quantifying uncertainty in neural networks.
method Showed deep ensembles perform exact Bayesian averaging with an implicitly learned data-dependent prior.
result Deep ensembles are Bayesian and provide an explanation for their strong empirical performance.
Bayesian inference is known to provide a general framework for incorporating prior knowledge or specific properties into machine learning models via carefully choosing a prior distribution. In this work, we propose a new type of prior distributions for convolutional neural networks, deep weight prior (DWP), that exploi…
BP-DIP combines DIP and backprojection for image restoration.
problem Performance drop of deep learning methods due to mismatched observation models.
method Combines Deep Image Prior (DIP) and backprojection.
result BP-DIP outperforms DIP in deblurring tasks with higher PSNR and faster inference.
PAC-Bayes bound requires prior to place mass on high-performing predictors.
problem Explaining generalization in machine learning.
method Analyzing necessary conditions for PAC-Bayes bounds to provide meaningful generalization guarantees.
result Achieving a target generalisation level requires the prior to place sufficient mass on high-performing predictors.
Proposes DAPr framework to learn feature importance from prior knowledge.
problem Ensuring meaningful feature attributions in deep models.
method Jointly learns feature importance from prior knowledge and biases models to rely on important features.
result Improves model generalization and provides new interpretation methods.
New theory for BNNs with Gaussian priors achieves optimal posterior concentration rates.
problem Lack of theoretical results for BNNs with Gaussian priors.
method New approximation theory for non-sparse DNNs with bounded parameters.
result BNNs with non-sparse general priors can achieve near-minimax optimal posterior concentration rates.
New algorithm uses untrained neural networks for image recovery, offering better compression.
problem Using untrained neural networks for image recovery and theoretical guarantees.
method Projected gradient descent scheme for solving linear and non-linear inverse problems.
result The method achieves better compression rates for the same image quality compared to hand-crafted priors.
Deep Gaussian processes can have non-degenerate and non-Gaussian limits.
problem Understanding the behavior of deep Gaussian processes as depth grows.
method Studying the limit of compositional Gaussian processes where each layer is a Gaussian process.
result Identified a sharp bandwidth threshold above which the limit is degenerate, and proved that for bandwidths below this threshold, the limit is a non-degenerate and non-Gaussian distribution.
New priors for deep neural networks converge to Gaussian processes.
problem Improving the performance and stability of deep neural networks.
method Extending prior distributions to include non-zero means and partially exchangeable priors, leading to a new Gaussian process model.
result The new Gaussian process model avoids pathologies and improves performance on regression problems.
New method learns priors for Bayesian neural networks from datasets.
problem Lack of prior beliefs in Bayesian deep learning.
method Amortised variational inference to learn priors from datasets.
result Flexible Bayesian neural networks for meta-learning and within-task minibatching.
The recent literature on deep learning offers new tools to learn a rich probability distribution over high dimensional data such as images or sounds. In this work we investigate the possibility of learning the prior distribution over neural network parameters using such tools. Our resulting variational Bayes algorithm …
Bayesian deep learning with heavy-tailed weights achieves near-optimal performance.
problem Deep neural networks with heavy-tailed weights achieve near-optimal performance in various contexts.
method Introduced a Bayesian deep learning prior based on heavy-tailed weights and ReLU activation, showing near-optimal minimax contraction rates.
result Posterior distribution achieves near-optimal minimax contraction rates, adaptive to smoothness and intrinsic dimension.
PriorCVAE uses deep generative models to infer hyperparameters in MCMC.
problem Losing hyperparameter information in GP prior inference.
method Conditioning VAE on hyperparameters to encode and estimate them during inference.
result PriorCVAE enables efficient and distinct inference of hyperparameters.
Proposes DAK model for improved GP computations.
problem Challenges in high-dimensional GP layers in DKL.
method Additive structure and induced prior approximation for GP units.
result Outperforms state-of-the-art DKL methods in regression and classification.
Proposes learning task-agnostic dynamics priors for faster RL.
problem Challenges in learning accurate dynamics models for RL.
method Pre-training a frame predictor on physics videos to initialize and fine-tune dynamics models.
result Improves policy learning and convergence, outperforming competitors.
New findings suggest latent regularization is unnecessary for high-quality image generation.
problem Improving image generation quality without latent regularization.
method Investigated the effect of latent regularization on image generation using learned priors.
result In the case of a sufficiently expressive prior, latent regularization is not necessary and may harm image quality.
RPPs improve deep learning models with soft equivariance constraints.
problem Balancing expressiveness and inductive biases in deep learning.
method Introducing Residual Pathway Priors (RPPs) to convert hard constraints into soft priors.
result RPPs enable models to learn structured solutions while retaining flexibility.
Bayesian deep learning faces posterior collapse due to likelihood vs. prior competition.
problem Posterior collapse in Bayesian deep learning models.
method Identified competition between likelihood and prior regularization in a linear latent variable model.
result Posterior collapse is related to neural and dimensional collapse, suggesting a broader learning issue.
CSGM framework applied to clinical MRI data for robust reconstructions.
problem Applying deep generative priors to clinical MRI data for high-quality reconstructions.
method Training a generative prior on brain scans from the fastMRI dataset and using Langevin dynamics for posterior sampling.
result Posterior sampling via Langevin dynamics achieves high quality reconstructions in clinical MRI data.
A new method for deep Wishart processes improves kernel-based models.
problem Inference in deep Wishart processes is challenging due to the need for flexible distributions over positive semi-definite matrices.
method Developed a novel approach to flexible distributions over positive semi-definite matrices using the Bartlett decomposition of the Wishart probability density. Used this to create an approximate posterior for the DWP.
result Improved performance of inference in the DWP compared to DGP with equivalent prior.
Posterior sampling estimator achieves near-optimal recovery guarantees for signals from any prior distribution.
problem Characterizing measurement complexity for signals from any prior distribution, including the entire space.
method Characterization of measurement complexity using posterior sampling estimator for Gaussian measurements and any prior distribution.
result Posterior sampling estimator achieves near-optimal recovery guarantees for signals from any prior distribution, robust to model mismatch.
Bayesian inference for deep neural networks using trace-class priors and MLMC.
problem Efficient Bayesian inference for deep neural networks.
method Trace-class neural network priors and Multilevel Monte Carlo method.
result Optimal computational complexity for Bayesian inference of TNN models.