A survey on epistemic (model) uncertainty in supervised learning: Recent advances and applications

X Zhou, H Liu, F Pourpanah, T Zeng, X Wang - Neurocomputing, 2022 - Elsevier
Quantifying the uncertainty of supervised learning models plays an important role in making
more reliable predictions. Epistemic uncertainty, which usually is due to insufficient …

Artificial intelligence (AI)—it's the end of the tox as we know it (and I feel fine)

N Kleinstreuer, T Hartung - Archives of Toxicology, 2024 - Springer
The rapid progress of AI impacts diverse scientific disciplines, including toxicology, and has
the potential to transform chemical safety evaluation. Toxicology has evolved from an …

Uncertainty quantification in scientific machine learning: Methods, metrics, and comparisons

AF Psaros, X Meng, Z Zou, L Guo… - Journal of Computational …, 2023 - Elsevier
Neural networks (NNs) are currently changing the computational paradigm on how to
combine data with mathematical laws in physics and engineering in a profound way …

Laplace redux-effortless bayesian deep learning

E Daxberger, A Kristiadi, A Immer… - Advances in …, 2021 - proceedings.neurips.cc
Bayesian formulations of deep learning have been shown to have compelling theoretical
properties and offer practical functional benefits, such as improved predictive uncertainty …

Do bayesian neural networks need to be fully stochastic?

M Sharma, S Farquhar, E Nalisnick… - International …, 2023 - proceedings.mlr.press
We investigate the benefit of treating all the parameters in a Bayesian neural network
stochastically and find compelling theoretical and empirical evidence that this standard …

Sampling from Gaussian process posteriors using stochastic gradient descent

JA Lin, J Antorán, S Padhy, D Janz… - Advances in …, 2023 - proceedings.neurips.cc
Gaussian processes are a powerful framework for quantifying uncertainty and for sequential
decision-making but are limited by the requirement of solving linear systems. In general, this …

Adapting the linearised laplace model evidence for modern deep learning

J Antorán, D Janz, JU Allingham… - International …, 2022 - proceedings.mlr.press
The linearised Laplace method for estimating model uncertainty has received renewed
attention in the Bayesian deep learning community. The method provides reliable error bars …

Bayesian low-rank adaptation for large language models

AX Yang, M Robeyns, X Wang, L Aitchison - arxiv preprint arxiv …, 2023 - arxiv.org
Parameter-efficient fine-tuning (PEFT) has emerged as a new paradigm for cost-efficient fine-
tuning of large language models (LLMs), with low-rank adaptation (LoRA) being a widely …

Gflowout: Dropout with generative flow networks

D Liu, M Jain, BFP Dossou, Q Shen… - International …, 2023 - proceedings.mlr.press
Bayesian inference offers principled tools to tackle many critical problems with modern
neural networks such as poor calibration and generalization, and data inefficiency …

Position paper: Bayesian deep learning in the age of large-scale ai

T Papamarkou, M Skoularidou, K Palla… - arxiv e …, 2024 - ui.adsabs.harvard.edu
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 …