About: Probabilistic Neighbourhood Component Analysis: Sample Efficient Uncertainty Estimation in Deep Learning

Facets (new session)
Description
Metadata
Settings
- owl:sameAs
- Inference Rule:

About: Probabilistic Neighbourhood Component Analysis: Sample Efficient Uncertainty Estimation in Deep Learning Goto Sponge NotDistinct Permalink

An Entity of Type : schema:ScholarlyArticle, within Data Space : covidontheweb.inria.fr associated with source document(s)

Attributes	Values
type	Academic Article research paper schema:ScholarlyArticle
isDefinedBy	Covid-on-the-Web dataset
title	Probabilistic Neighbourhood Component Analysis: Sample Efficient Uncertainty Estimation in Deep Learning
Creator	Dwivedi, Chaitanya Han, Yong-Jin Joshi, Gauri Kailkhura, Bhavya Mallick, Ankur
source	ArXiv
abstract	While Deep Neural Networks (DNNs) achieve state-of-the-art accuracy in various applications, they often fall short in accurately estimating their predictive uncertainty and, in turn, fail to recognize when these predictions may be wrong. Several uncertainty-aware models, such as Bayesian Neural Network (BNNs) and Deep Ensembles have been proposed in the literature for quantifying predictive uncertainty. However, research in this area has been largely confined to the big data regime. In this work, we show that the uncertainty estimation capability of state-of-the-art BNNs and Deep Ensemble models degrades significantly when the amount of training data is small. To address the issue of accurate uncertainty estimation in the small-data regime, we propose a probabilistic generalization of the popular sample-efficient non-parametric kNN approach. Our approach enables deep kNN classifier to accurately quantify underlying uncertainties in its prediction. We demonstrate the usefulness of the proposed approach by achieving superior uncertainty quantification as compared to state-of-the-art on a real-world application of COVID-19 diagnosis from chest X-Rays. Our code is available at https://github.com/ankurmallick/sample-efficient-uq
has issue date	2020-07-18(xsd:dateTime)
has license	arxiv
sha1sum (hex)	3aff58b3a6db8fb50c63eb0e78efad97326d6211
resource representing a document's title	Probabilistic Neighbourhood Component Analysis: Sample Efficient Uncertainty Estimation in Deep Learning
resource representing a document's body	covid:3aff58b3a6db8fb50c63eb0e78efad97326d6211#body_text
is schema:about of	named entity 'Our' named entity 'Neural Network' named entity 'RESEARCH' named entity 'UNCERTAINTY' named entity 'ITS' named entity 'PREDICTION' named entity 'ACCURACY' named entity 'RECOGNIZE' named entity 'OF-' named entity 'ADDRESS' named entity 'KNN' named entity 'EFFICIENT' named entity 'UNCERTAINTY' named entity 'COMPONENT' named entity 'recognize' named entity 'short' named entity 'confined' named entity 'predictive' named entity 'compared' named entity 'popular' named entity 'X-Rays' named entity 'github.com' named entity 'uncertainty' named entity 'Uncertainty' named entity 'Ensemble models' named entity 'state-of-the-art' named entity 'uncertainty quantification' named entity 'non-parametric' named entity 'COVID-19' named entity 'Component Analysis' named entity 'data point' named entity 'latent space' named entity 'probability' named entity 'latent space' named entity 'TensorFlow' named entity 'neural network' named entity 'training data' named entity 'PNCA' named entity 'DNN' named entity 'Reproducing Kernel Hilbert Space' named entity 'MNIST' named entity 'Orthogonal' named entity 'CXR' named entity 'non-parametric' named entity 'PNCA' named entity 'probability distributions' named entity 'data point' named entity 'algorithm' named entity 'training dataset' named entity 'chest X-Ray' named entity 'BNN' named entity 'data point' named entity 'RKHS' named entity 'RKHS' named entity 'stochasticity' named entity 'X-Ray' named entity 'probability' named entity 'latent space' named entity 'gradient descent' named entity 'COVID-19' named entity 'X-Ray' named entity 'BNN' named entity 'CXR' named entity 'deep learning' named entity 'MNIST' named entity 'training dataset' named entity 'RKHS' named entity 'PNCA' named entity 'data samples' named entity 'BNN' named entity 'overfit' named entity 'training data' named entity 'chest X-Ray' named entity 'classifier' named entity 'COVID-19' named entity 'training data' named entity 'COVID-19' named entity 'data classification' named entity 'probability' named entity 'COVID-19' named entity 'distance metric' named entity 'uncertainty quantification' named entity 'training data'

Faceted Search & Find service v1.13.91 as of Mar 24 2020

Alternative Linked Data Documents: Sponger | ODE Content Formats:

RDF

ODATA

Microdata

About

OpenLink Virtuoso version 07.20.3229 as of Jul 10 2020, on Linux (x86_64-pc-linux-gnu), Single-Server Edition (94 GB total memory)
Data on this page belongs to its respective rights holders.
Virtuoso Faceted Browser Copyright © 2009-2025 OpenLink Software