BayesCap: Bayesian Identity Cap for Calibrated Uncertainty in Frozen
Neural Networks
- URL: http://arxiv.org/abs/2207.06873v1
- Date: Thu, 14 Jul 2022 12:50:09 GMT
- Title: BayesCap: Bayesian Identity Cap for Calibrated Uncertainty in Frozen
Neural Networks
- Authors: Uddeshya Upadhyay, Shyamgopal Karthik, Yanbei Chen, Massimiliano
Mancini, Zeynep Akata
- Abstract summary: We propose BayesCap that learns a Bayesian identity mapping for the frozen model, allowing uncertainty estimation.
BayesCap is a memory-efficient method that can be trained on a small fraction of the original dataset.
We show the efficacy of our method on a wide variety of tasks with a diverse set of architectures.
- Score: 50.15201777970128
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: High-quality calibrated uncertainty estimates are crucial for numerous
real-world applications, especially for deep learning-based deployed ML
systems. While Bayesian deep learning techniques allow uncertainty estimation,
training them with large-scale datasets is an expensive process that does not
always yield models competitive with non-Bayesian counterparts. Moreover, many
of the high-performing deep learning models that are already trained and
deployed are non-Bayesian in nature and do not provide uncertainty estimates.
To address these issues, we propose BayesCap that learns a Bayesian identity
mapping for the frozen model, allowing uncertainty estimation. BayesCap is a
memory-efficient method that can be trained on a small fraction of the original
dataset, enhancing pretrained non-Bayesian computer vision models by providing
calibrated uncertainty estimates for the predictions without (i) hampering the
performance of the model and (ii) the need for expensive retraining the model
from scratch. The proposed method is agnostic to various architectures and
tasks. We show the efficacy of our method on a wide variety of tasks with a
diverse set of architectures, including image super-resolution, deblurring,
inpainting, and crucial application such as medical image translation.
Moreover, we apply the derived uncertainty estimates to detect
out-of-distribution samples in critical scenarios like depth estimation in
autonomous driving. Code is available at
https://github.com/ExplainableML/BayesCap.
Related papers
- Estimating Uncertainty with Implicit Quantile Network [0.0]
Uncertainty quantification is an important part of many performance critical applications.
This paper provides a simple alternative to existing approaches such as ensemble learning and bayesian neural networks.
arXiv Detail & Related papers (2024-08-26T13:33:14Z) - Large Language Models Must Be Taught to Know What They Don't Know [97.90008709512921]
We show that fine-tuning on a small dataset of correct and incorrect answers can create an uncertainty estimate with good generalization and small computational overhead.
We also investigate the mechanisms that enable reliable uncertainty estimation, finding that many models can be used as general-purpose uncertainty estimators.
arXiv Detail & Related papers (2024-06-12T16:41:31Z) - Post-hoc Uncertainty Learning using a Dirichlet Meta-Model [28.522673618527417]
We propose a novel Bayesian meta-model to augment pre-trained models with better uncertainty quantification abilities.
Our proposed method requires no additional training data and is flexible enough to quantify different uncertainties.
We demonstrate our proposed meta-model approach's flexibility and superior empirical performance on these applications.
arXiv Detail & Related papers (2022-12-14T17:34:11Z) - NUQ: Nonparametric Uncertainty Quantification for Deterministic Neural
Networks [151.03112356092575]
We show the principled way to measure the uncertainty of predictions for a classifier based on Nadaraya-Watson's nonparametric estimate of the conditional label distribution.
We demonstrate the strong performance of the method in uncertainty estimation tasks on a variety of real-world image datasets.
arXiv Detail & Related papers (2022-02-07T12:30:45Z) - $f$-Cal: Calibrated aleatoric uncertainty estimation from neural
networks for robot perception [9.425514903472545]
Existing approaches estimate uncertainty from neural network perception stacks by modifying network architectures, inference procedure, or loss functions.
Our key insight is that calibration is only achieved by imposing constraints across multiple examples, such as those in a mini-batch.
By enforcing the distribution of outputs of a neural network to resemble a target distribution by minimizing an $f$-divergence, we obtain significantly better-calibrated models compared to prior approaches.
arXiv Detail & Related papers (2021-09-28T17:57:58Z) - PDC-Net+: Enhanced Probabilistic Dense Correspondence Network [161.76275845530964]
Enhanced Probabilistic Dense Correspondence Network, PDC-Net+, capable of estimating accurate dense correspondences.
We develop an architecture and an enhanced training strategy tailored for robust and generalizable uncertainty prediction.
Our approach obtains state-of-the-art results on multiple challenging geometric matching and optical flow datasets.
arXiv Detail & Related papers (2021-09-28T17:56:41Z) - Understanding Uncertainty in Bayesian Deep Learning [0.0]
We show that traditional training procedures for NLMs can drastically underestimate uncertainty in data-scarce regions.
We propose a novel training method that can both capture useful predictive uncertainties as well as allow for incorporation of domain knowledge.
arXiv Detail & Related papers (2021-05-21T19:22:17Z) - Unlabelled Data Improves Bayesian Uncertainty Calibration under
Covariate Shift [100.52588638477862]
We develop an approximate Bayesian inference scheme based on posterior regularisation.
We demonstrate the utility of our method in the context of transferring prognostic models of prostate cancer across globally diverse populations.
arXiv Detail & Related papers (2020-06-26T13:50:19Z) - Uncertainty-Aware (UNA) Bases for Deep Bayesian Regression Using
Multi-Headed Auxiliary Networks [23.100727871427367]
We show that traditional training procedures for Neural Linear Models drastically underestimate uncertainty on out-of-distribution inputs.
We propose a novel training framework that captures useful predictive uncertainties for downstream tasks.
arXiv Detail & Related papers (2020-06-21T02:46:05Z) - Diversity inducing Information Bottleneck in Model Ensembles [73.80615604822435]
In this paper, we target the problem of generating effective ensembles of neural networks by encouraging diversity in prediction.
We explicitly optimize a diversity inducing adversarial loss for learning latent variables and thereby obtain diversity in the output predictions necessary for modeling multi-modal data.
Compared to the most competitive baselines, we show significant improvements in classification accuracy, under a shift in the data distribution.
arXiv Detail & Related papers (2020-03-10T03:10:41Z)
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.