Gradient-similarity-regularized weight averaging and WA+SAM fine-tuning are tested on OOD and few-shot domain adaptation benchmarks, with mixed results that do not support the claimed improvements.
Robustness to corruption in pre-trained Bayesian neural networks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
We develop ShiftMatch, a new training-data-dependent likelihood for robustness to corruption in Bayesian neural networks (BNNs). ShiftMatch is inspired by the training-data-dependent "EmpCov" priors from Izmailov et al. (2021a), and efficiently matches test-time spatial correlations to those at training time. Critically, ShiftMatch is designed to leave the neural network's training time likelihood unchanged, allowing it to use publicly available samples from pre-trained BNNs. Using pre-trained HMC samples, ShiftMatch gives strong performance improvements on CIFAR-10-C, outperforms EmpCov priors (though ShiftMatch uses extra information from a minibatch of corrupted test points), and is perhaps the first Bayesian method capable of convincingly outperforming plain deep ensembles.
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
REJECT 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Weight Averaging for Out-of-Distribution Generalization and Few-Shot Domain Adaptation
Gradient-similarity-regularized weight averaging and WA+SAM fine-tuning are tested on OOD and few-shot domain adaptation benchmarks, with mixed results that do not support the claimed improvements.