Bayesian Neural Network Priors Revisited

• URL: http://arxiv.org/abs/2102.06571v1
• Date: Fri, 12 Feb 2021 15:18:06 GMT
• Title: Bayesian Neural Network Priors Revisited
• Authors: Vincent Fortuin, Adri\`a Garriga-Alonso, Florian Wenzel, Gunnar R\"atsch, Richard Turner, Mark van der Wilk, Laurence Aitchison
• Abstract summary: We study summary statistics of neural network weights in different networks trained using SGD. We find that fully connected networks (FCNNs) display heavy-tailed weight distributions, while convolutional neural network (CNN) weights display strong spatial correlations.
• Score: 29.949163519715952
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Isotropic Gaussian priors are the de facto standard for modern Bayesian neural network inference. However, such simplistic priors are unlikely to either accurately reflect our true beliefs about the weight distributions, or to give optimal performance. We study summary statistics of neural network weights in different networks trained using SGD. We find that fully connected networks (FCNNs) display heavy-tailed weight distributions, while convolutional neural network (CNN) weights display strong spatial correlations. Building these observations into the respective priors leads to improved performance on a variety of image classification datasets. Moreover, we find that these priors also mitigate the cold posterior effect in FCNNs, while in CNNs we see strong improvements at all temperatures, and hence no reduction in the cold posterior effect.

Related papers

• Deep Neural Networks Tend To Extrapolate Predictably [51.303814412294514]
  neural network predictions tend to be unpredictable and overconfident when faced with out-of-distribution (OOD) inputs. We observe that neural network predictions often tend towards a constant value as input data becomes increasingly OOD. We show how one can leverage our insights in practice to enable risk-sensitive decision-making in the presence of OOD inputs.
  arXiv  Detail & Related papers  (2023-10-02T03:25:32Z)
• Spike-and-slab shrinkage priors for structurally sparse Bayesian neural networks [0.16385815610837165]
  Sparse deep learning addresses challenges by recovering a sparse representation of the underlying target function. Deep neural architectures compressed via structured sparsity provide low latency inference, higher data throughput, and reduced energy consumption. We propose structurally sparse Bayesian neural networks which prune excessive nodes with (i) Spike-and-Slab Group Lasso (SS-GL), and (ii) Spike-and-Slab Group Horseshoe (SS-GHS) priors.
  arXiv  Detail & Related papers  (2023-08-17T17:14:18Z)
• Addressing caveats of neural persistence with deep graph persistence [54.424983583720675]
  We find that the variance of network weights and spatial concentration of large weights are the main factors that impact neural persistence. We propose an extension of the filtration underlying neural persistence to the whole neural network instead of single layers. This yields our deep graph persistence measure, which implicitly incorporates persistent paths through the network and alleviates variance-related issues.
  arXiv  Detail & Related papers  (2023-07-20T13:34:11Z)
• Sparsifying Bayesian neural networks with latent binary variables and normalizing flows [10.865434331546126]
  We will consider two extensions to the latent binary Bayesian neural networks (LBBNN) method. Firstly, by using the local reparametrization trick (LRT) to sample the hidden units directly, we get a more computationally efficient algorithm. More importantly, by using normalizing flows on the variational posterior distribution of the LBBNN parameters, the network learns a more flexible variational posterior distribution than the mean field Gaussian.
  arXiv  Detail & Related papers  (2023-05-05T09:40:28Z)
• Gradient Descent in Neural Networks as Sequential Learning in RKBS [63.011641517977644]
  We construct an exact power-series representation of the neural network in a finite neighborhood of the initial weights. We prove that, regardless of width, the training sequence produced by gradient descent can be exactly replicated by regularized sequential learning.
  arXiv  Detail & Related papers  (2023-02-01T03:18:07Z)
• Wide Bayesian neural networks have a simple weight posterior: theory and accelerated sampling [48.94555574632823]
  Repriorisation transforms a Bayesian neural network (BNN) posterior to a distribution whose KL divergence to the BNN prior vanishes as layer widths grow. We develop a Markov chain Monte Carlo (MCMC) posterior sampling algorithm which mixes faster the wider the BNN. We observe up to 50x higher effective sample size relative to no reparametrisation for both fully-connected and residual networks.
  arXiv  Detail & Related papers  (2022-06-15T17:11:08Z)
• BNNpriors: A library for Bayesian neural network inference with different prior distributions [32.944046414823916]
  BNNpriors enables state-of-the-art Markov Chain Monte Carlo inference on Bayesian neural networks. It follows a modular approach that eases the design and implementation of new custom priors. It has facilitated foundational discoveries on the nature of the cold posterior effect in Bayesian neural networks.
  arXiv  Detail & Related papers  (2021-05-14T17:11:04Z)
• Stochastic Bayesian Neural Networks [0.0]
  We build on variational inference techniques for bayesian neural networks using the original Evidence Lower Bound. We present a bayesian neural network in which we maximize Evidence Lower Bound using a new objective function which we name as Evidence Lower Bound.
  arXiv  Detail & Related papers  (2020-08-12T19:48:34Z)
• Finite Versus Infinite Neural Networks: an Empirical Study [69.07049353209463]
  kernel methods outperform fully-connected finite-width networks. Centered and ensembled finite networks have reduced posterior variance. Weight decay and the use of a large learning rate break the correspondence between finite and infinite networks.
  arXiv  Detail & Related papers  (2020-07-31T01:57:47Z)
• Revisiting Initialization of Neural Networks [72.24615341588846]
  We propose a rigorous estimation of the global curvature of weights across layers by approximating and controlling the norm of their Hessian matrix. Our experiments on Word2Vec and the MNIST/CIFAR image classification tasks confirm that tracking the Hessian norm is a useful diagnostic tool.
  arXiv  Detail & Related papers  (2020-04-20T18:12:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.