REVIEW 7 cited by
Position: Bayesian Deep Learning is Needed in the Age of Large-Scale AI
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
In the current landscape of deep learning research, there is a predominant emphasis on achieving high predictive accuracy in supervised tasks involving large image and language datasets. However, a broader perspective reveals a multitude of overlooked metrics, tasks, and data types, such as uncertainty, active and continual learning, and scientific data, that demand attention. Bayesian deep learning (BDL) constitutes a promising avenue, offering advantages across these diverse settings. This paper posits that BDL can elevate the capabilities of deep learning. It revisits the strengths of BDL, acknowledges existing challenges, and highlights some exciting research avenues aimed at addressing these obstacles. Looking ahead, the discussion focuses on possible ways to combine large-scale foundation models with BDL to unlock their full potential.
Forward citations
Cited by 7 Pith papers
-
laplax -- Laplace Approximations with JAX
The paper presents laplax, a modular JAX library for Laplace approximations that supports multiple curvature estimates, uncertainty pushforwards, calibration, and evaluation routines.
-
Bayesian Deep Learning for Discrete Choice
A Bayesian deep learning architecture for discrete choice yields calibrated credible intervals for marginal rates of substitution and collapses to linear utility models when data are limited.
-
Scalable Bayesian Monte Carlo: fast uncertainty estimation beyond deep ensembles
SBMC samples from a MAP-anchored approximate posterior with parallel MCMC/SMC chains, delivering deep-ensemble-level accuracy with better calibrated epistemic uncertainty at similar cost.
-
Approximate Message Passing for Bayesian Neural Networks
A factor-graph message-passing method for Bayesian neural networks that handles CNNs, avoids double-counting, and shows competitive accuracy with improved calibration on CIFAR-10.
-
Single- to multi-fidelity history-dependent learning with uncertainty quantification and disentanglement: application to data-driven constitutive modeling
A multi-fidelity Bayesian recurrent neural network framework predicts history-dependent material responses while separately quantifying aleatoric and epistemic uncertainties.
-
Uncertainty in Physics and AI: Taxonomy, Quantification, and Validation
A unified taxonomy of uncertainty in ML for physics is introduced together with validation tools such as coverage, calibration, and proper scoring rules, illustrated on regression and classification tasks.
-
Why you don't overfit, and don't need Bayes if you only train for one epoch
In one-epoch, data-rich training, minibatch maximum likelihood is an unbiased estimator of the test loss, which the paper claims makes Bayesian inference unnecessary.
Discussion (0). Continue with ORCID to comment.